Maintenance for the week of December 23:
· [COMPLETE] NA megaservers for maintenance – December 23, 4:00AM EST (9:00 UTC) - 9:00AM EST (14:00 UTC)
· [COMPLETE] EU megaservers for maintenance – December 23, 9:00 UTC (4:00AM EST) - 14:00 UTC (9:00AM EST)

Procedurally Generated Dungeons?

Calsifer
Calsifer
✭✭✭
Hey Dev Team,

I was just curious, but has any thought been put into a procedurally generated dungeon to shake things up as it were? Much like how in the early days of a game with a certain "El Diablo" in which the levels were always different each time you played them?

This would break up the monotonous nature of running random (seen it, been there, done that) dungeons.

My thoughs would be along the lines of a dungeon that has random map layouts, and bosses (with random abilities). This would blind side every group each time the dungeon is run. You would never know what to expect each run and could never anticipate exactly what skills would need to be brought to the table.

Well balanced characters might take a bit longer to get through the dungeon, but so called "glass cannons" might not have the tool sets necessary to just BLAST their way through it. Perhaps skills based on the class/role you chose would make suitable obstacles/objectives possible, i.e. doors that require the highest level of lockpicking, or bottomless pits that require streak to get across, to then lower a bridge for non teleporting characters to get across.

Not necessarily making things puzzles. (personally I am too stupid to do puzzles on the fly) But adding new content constantly without the need for a DLC or at least something to truly hold us over and give us an ADVENTURE rather than simply the normal dungeon thrashing that happens now.

Individual mechanics could be learned, but not anticipated for each run. The fire runes on the floor the first time you run around the 3rd corder may be a boss next time or trash the following time, or even a dead end. (as the map should be constantly changing with each new run) Not many new textures would be required, as with Skyrim, while many caves look to have exactly the same textures and a few similar rooms, they had different layouts using the same "tool set" so to speak.

I believe this proposition would allow the dev team more time to bring QUALITY DLCs and chapters.
  • Arobain
    Arobain
    ✭✭✭✭
    i like this idea alot, like seriously
  • Peekachu99
    Peekachu99
    ✭✭✭✭✭
    ✭✭
    WoW is doing this in their next Expac—expect others to follow. Would be great if ESO gave it a whirl, too.
  • ArchMikem
    ArchMikem
    ✭✭✭✭✭
    ✭✭✭✭✭
    Not intending to bash the Dev team, but if the Group Finder alone can't seem to work right how'd you suppose entire randomly generated Dungeons would fare?
    CP2,000 Master Explorer - AvA One Star General - Console Peasant - The Clan
    Quest Objective: OMG Go Talk To That Kitty!
  • Calsifer
    Calsifer
    ✭✭✭
    Blizzard may have a leg up in this idea as they did it in . . . Diablo. The first one at least. But, having said that. ESO could do it Much MUCH better.
  • Calsifer
    Calsifer
    ✭✭✭
    I believe in the power of this tool. Sure there are bugs, but if they were not working on new content CONSTANTLY, perhaps they would have more time to squash bugs. huh? maybe? ;)
  • Calsifer
    Calsifer
    ✭✭✭
    This idea does not even need to be limited to 4 man groups. I have heard whispers of folks asking for a new "arena" like activity. Perhaps a Solo version of this would work as well?

    You would need a very well balanced character to get past the obstacles. Perhaps driving players to oh say, play the game, to get enough skill points to be able to have all the abilities on one character.

    This in turn would give players more incentive to do main story questing while waiting for new content to be released. I know many players that have max level/max CP toons with almost NONE of the story quests completed.
  • Rainraven
    Rainraven
    ✭✭✭✭✭
    Yes please.
  • VaranisArano
    VaranisArano
    ✭✭✭✭✭
    ✭✭✭✭✭
    Oh, like in TES IV: Oblivion with the Oblivion gates?

    Yeah, that didn't get boring really quick.
  • LadyLethalla
    LadyLethalla
    ✭✭✭✭✭
    Oh, like in TES IV: Oblivion with the Oblivion gates?

    Yeah, that didn't get boring really quick.

    Clearly those were limited to two or three locations. Just as with Skyrim - look how many of the dungeons and caves are the same. What Calsifer is asking is that even if it's the same dungeon layout(s) the mobs/bosses contained therein are different. Little changed inside the Oblivion gates from what I remember. Skyrim (which I'm currently playing on VR) at least had a bit of variety. Don't forget as well, how much games have expanded since Oblivion's release (2006). You can't compare them.

    Personally I like the idea of people not being able to learn the dungeon well enough that they could do it in their sleep.
    Would also like that if players are able to faceroll the boss that it gets progressively more difficult.


    Edited by LadyLethalla on December 19, 2017 3:03AM
    x-TallyCat-x // PC EU DC - For the Covenant! // ESO Platinum trophy - 16th May 2017.
    Melbourne Australia - the land of Potato Internet.WTB ESO OCEANIC SERVER
  • mikemacon
    mikemacon
    ✭✭✭✭✭
    Calsifer wrote: »
    Hey Dev Team,

    I was just curious, but has any thought been put into a procedurally generated dungeon to shake things up as it were? Much like how in the early days of a game with a certain "El Diablo" in which the levels were always different each time you played them?

    This would break up the monotonous nature of running random (seen it, been there, done that) dungeons.

    My thoughs would be along the lines of a dungeon that has random map layouts, and bosses (with random abilities). This would blind side every group each time the dungeon is run. You would never know what to expect each run and could never anticipate exactly what skills would need to be brought to the table.

    Well balanced characters might take a bit longer to get through the dungeon, but so called "glass cannons" might not have the tool sets necessary to just BLAST their way through it. Perhaps skills based on the class/role you chose would make suitable obstacles/objectives possible, i.e. doors that require the highest level of lockpicking, or bottomless pits that require streak to get across, to then lower a bridge for non teleporting characters to get across.

    Not necessarily making things puzzles. (personally I am too stupid to do puzzles on the fly) But adding new content constantly without the need for a DLC or at least something to truly hold us over and give us an ADVENTURE rather than simply the normal dungeon thrashing that happens now.

    Individual mechanics could be learned, but not anticipated for each run. The fire runes on the floor the first time you run around the 3rd corder may be a boss next time or trash the following time, or even a dead end. (as the map should be constantly changing with each new run) Not many new textures would be required, as with Skyrim, while many caves look to have exactly the same textures and a few similar rooms, they had different layouts using the same "tool set" so to speak.

    I believe this proposition would allow the dev team more time to bring QUALITY DLCs and chapters.

    YES PLEASE...
  • Florial
    Florial
    ✭✭✭
    Wasn't this done in Daggerfall? I recall being lost on some multidimensional dungeons in some Elder Scrolls game.
    Edited by Florial on December 19, 2017 3:58AM
  • Calsifer
    Calsifer
    ✭✭✭
    Just brainstorming here, but on the Solo run of such a dungeon, it may simply not be possible to complete if you lack a balance of skills. Still it would not simply be a DPS race.

    I for one live in Okinawa Japan, and play on PS4 EU server. I have horroble connectivity to the servers most of the time, and having some other method of achievement asside from animation canceling and weaving/rotations would be nice.
  • Appleblade
    Appleblade
    ✭✭✭✭✭
    Maybe add some new procedural delves to each zone as a start.

    I’ve dabbled in procedural generation. The trick isn’t generating the probabilistic structure, it’s having enough unique assets to populate the structure so it doesn’t get repetitive. That was No Man’s Sky problem. It doesn’t matter if you have 50 gigabazillion planets if you start seeing the same eight trees and eleven rocks over and over again. It’s a solvable problem. You just have to push the randomness further into the details. I would’ve had the trees be fractal structures, for example. Even leaf shapes would’ve been procedural .

    Edited by Appleblade on December 19, 2017 7:17AM
  • mocap
    mocap
    ✭✭✭✭✭
    Someone play Path of Exile too much )
  • Doctordarkspawn
    Doctordarkspawn
    ✭✭✭✭✭
    ✭✭✭✭✭
    Now, I've seen talk of this before, and if they can get the system in place, I think it'd be cool and a great way to enhance the dungeon experience.

    The only problem would be setting it up.
  • Calsifer
    Calsifer
    ✭✭✭
    mocap wrote: »
    Someone play Path of Exile too much )

    Honestly, I basically only play ESO, and would like to see more of a reason for MORE people to only play ESO. ;)

  • OmniDo
    OmniDo
    ✭✭✭✭✭
    This is a very, very, old idea.
    Three reasons why it wont happen in the next decade:
    1. Far too many variables (at present) to code for a stable, bug free result
    2. Finite amount of database space to store variations
    3. Too much computation time required for current server technology, given the cost/profit ratio
    But its been a very old idea, suggested by many to various developers of various games and MMO's, some directly to developer teams (such as myself to Turbine, back when DDO was in early development) with the same responses.
    Quite simply, it would be a huge R&D expenditure with no guaranteed result, since no one in the industry has done it.

    I will say that the first to successfully achieve it though will /win the game of endless quality entertainment, as all current games require a creative human element to them that demands <x> amount of time investment for a static product.
    Will it happen?
    Eventually, yes.
    The technology exists today to support it, its just something that has never been successfully implemented.
    Edited by OmniDo on December 19, 2017 8:00AM
  • Ackwalan
    Ackwalan
    ✭✭✭✭✭
    ✭✭✭✭✭
    This might work for group dungeons that are instanced to that group, but I don't think it is possible for delves and public dungeons.
  • Darkstorne
    Darkstorne
    ✭✭✭✭✭
    ✭✭
    OmniDo wrote: »
    This is a very, very, old idea.
    Three reasons why it wont happen in the next decade:
    1. Far too many variables (at present) to code for a stable, bug free result
    2. Finite amount of database space to store variations
    3. Too much computation time required for current server technology, given the cost/profit ratio
    But its been a very old idea, suggested by many to various developers of various games and MMO's, some directly to developer teams (such as myself to Turbine, back when DDO was in early development) with the same responses.
    Quite simply, it would be a huge R&D expenditure with no guaranteed result, since no one in the industry has done it.

    I will say that the first to successfully achieve it though will /win the game of endless quality entertainment, as all current games require a creative human element to them that demands <x> amount of time investment for a static product.
    Will it happen?
    Eventually, yes.
    The technology exists today to support it, its just something that has never been successfully implemented.

    What are you talking about? O.o

    Diablo 3 is the perfect example of why this would be an amazing idea. They handled it incredibly well, and WoW is implementing a similar design for its next expac. It's the perfect concept for the next set of challenge weapons added to ESO, and would add an immense amount of new gameplay for both solo and group players if executed well. Also, once the system in place, they should be able to add new tilesets for different environments with relative ease.

    I suggested this a couple of months back and mentioned deep Valenwood would be perfect for a procedural dungeon design that changes layout with each visit. Moving trees and all that.
  • Doctordarkspawn
    Doctordarkspawn
    ✭✭✭✭✭
    ✭✭✭✭✭
    OmniDo wrote: »
    This is a very, very, old idea.
    Three reasons why it wont happen in the next decade:
    1. Far too many variables (at present) to code for a stable, bug free result
    2. Finite amount of database space to store variations
    3. Too much computation time required for current server technology, given the cost/profit ratio
    But its been a very old idea, suggested by many to various developers of various games and MMO's, some directly to developer teams (such as myself to Turbine, back when DDO was in early development) with the same responses.
    Quite simply, it would be a huge R&D expenditure with no guaranteed result, since no one in the industry has done it.

    I will say that the first to successfully achieve it though will /win the game of endless quality entertainment, as all current games require a creative human element to them that demands <x> amount of time investment for a static product.
    Will it happen?
    Eventually, yes.
    The technology exists today to support it, its just something that has never been successfully implemented.

    Yeah this is about what I expected.
  • OmniDo
    OmniDo
    ✭✭✭✭✭
    Darkstorne wrote: »
    What are you talking about? O.o
    Diablo 3 is the perfect example of why this would be an amazing idea. They handled it incredibly well, and WoW is implementing a similar design for its next expac.
    Diablo 3 is fun to be sure, but the "random" patterns are just the same ol' stuff, with minor variations.
    You dont really see anything "new".
    The goal of procedurally generated content is to accomplish two things:
    1. Eliminate pattern detection (humans are notoriously good at this)
    2. Facilitate near-infinite content based upon finite player decision, while simultaneously accomplishing #1
    Thus far, the only game I know of that has done this is Elite Dangerous, and not because its entirely procedurally random, but rather the developers modeled an entire galaxy within their explore-able space, and thats some 400 billion'ish star systems.
    It would take a single player over 1,095,890,410 years to explore every last unique star system, presuming they could both travel to, and completely catalog one star system every 24 hours. (400 billion / 365)
    Or, it would take 1,095,890,410 players 1 year, however you want to do the math.

    The best one could conceivably do is to model "Tropes" for story content and then a range of "scenarios" for interactive/reactive content.
    Since humans are very good at detecting specific repetitive patterns, only generic patterns (tropes & styles) are regarded as "Fresh" and new.
    Play w0w for any length of time and you have learned all the possible variations for all mob attacks, as well as becoming intimately familiar with the landscapes and having already mapped out the shortest routes from A to B.

    To eliminate this, a game has to have only a "Generic" atmosphere, i.e the area is "Bumpy" or "Desolate", in addition to generic challenges, "The Hounds of scorchland are no stranger to fire", without having specificity that would inevitably become "too familiar".
    With static content players get "bored", the same as they get bored with D3's "random" rifts, eventually.
    You can only maintain the "Monty Hall" campaign for so long before it simply gets old.
    Edited by OmniDo on December 19, 2017 9:22AM
  • Denyiir
    Denyiir
    ✭✭✭✭
    Reading about your idea instantly reminded me dragon age 2 :lol:
  • pandoraderomanus
    pandoraderomanus
    ✭✭✭
    Sounds like great idea. Reminds me of great portals in diablo3, you never know what is behind next 'door' which sometimes leads to epic fail even if you previously did the same level of portal. Would be also needed to add normal/veteran/hardcore difficulty and chest with some goods at the end.
    Can add there also troll levels, like uncle Sheo wants to play with you mortal, some annoying tricks with teleporting back to the beginning of the room or trying to make cheese from small Valkyn Skoria copy with wabbajack. Add levels from other daedra too, like Hircine or whoever was that werewolf prince will make you a werewolf and you need to kill everything in werewolf form.

    #moreskoomaforlizards
    PC-EU since 2014

    Touches-Your-Tralala - retired lizardina-templar
    Pandora Morgenstern - noob orc-stamsorc
  • altemriel
    altemriel
    ✭✭✭✭✭
    ✭✭✭
    Calsifer wrote: »
    Hey Dev Team,

    I was just curious, but has any thought been put into a procedurally generated dungeon to shake things up as it were? Much like how in the early days of a game with a certain "El Diablo" in which the levels were always different each time you played them?

    This would break up the monotonous nature of running random (seen it, been there, done that) dungeons.

    My thoughs would be along the lines of a dungeon that has random map layouts, and bosses (with random abilities). This would blind side every group each time the dungeon is run. You would never know what to expect each run and could never anticipate exactly what skills would need to be brought to the table.

    Well balanced characters might take a bit longer to get through the dungeon, but so called "glass cannons" might not have the tool sets necessary to just BLAST their way through it. Perhaps skills based on the class/role you chose would make suitable obstacles/objectives possible, i.e. doors that require the highest level of lockpicking, or bottomless pits that require streak to get across, to then lower a bridge for non teleporting characters to get across.

    Not necessarily making things puzzles. (personally I am too stupid to do puzzles on the fly) But adding new content constantly without the need for a DLC or at least something to truly hold us over and give us an ADVENTURE rather than simply the normal dungeon thrashing that happens now.

    Individual mechanics could be learned, but not anticipated for each run. The fire runes on the floor the first time you run around the 3rd corder may be a boss next time or trash the following time, or even a dead end. (as the map should be constantly changing with each new run) Not many new textures would be required, as with Skyrim, while many caves look to have exactly the same textures and a few similar rooms, they had different layouts using the same "tool set" so to speak.

    I believe this proposition would allow the dev team more time to bring QUALITY DLCs and chapters.



    this would be really good!!
  • FrancisCrawford
    FrancisCrawford
    ✭✭✭✭✭
    ✭✭✭✭
    Dolmens have bosses chosen at apparent random from a fixed list. Skirmishes in LOTRO worked similarly.

    That's a start.

  • zaria
    zaria
    ✭✭✭✭✭
    ✭✭✭✭✭
    As an starter how about an chaotic boss, 25 general boss abilities who is used on other bosses. he get say 3 to 5.
    Some lockout so he will not link two players like in FG2 and then pull them inn like FH and some others :)
    Grinding just make you go in circles.
    Asking ZoS for nerfs is as stupid as asking for close air support from the death star.
  • Thraben
    Thraben
    ✭✭✭✭✭
    Yes, there were skirmishes in Lotro with procedually generated mob groups and mini bosses (with like 50 or so different mechanics).

    Could be really fun at times.

    Lorewise, a raid on any daedric plane would fit, or a necromancer's layer.
    Hauptmann der Dolche des Königs

    DDK ist die letzte Verteidigungslinie des Dolchsturz- Bündnisses auf der 30-Tage-No-CP- Kampagne(EU) mit dem Anspruch, in kleinen, anfängerfreundlichen Raid-Gruppen möglichst epische Schlachten auszufechten.

    DDK is the Daggerfall Covenant´s last line of defense on the 30 days no-cp campaign (EU). We intend to fight epic battles in small, casual player friendly raid groups.
  • Calsifer
    Calsifer
    ✭✭✭
    OmniDo wrote: »
    This is a very, very, old idea.
    Three reasons why it wont happen in the next decade:
    1. Far too many variables (at present) to code for a stable, bug free result
    2. Finite amount of database space to store variations
    3. Too much computation time required for current server technology, given the cost/profit ratio
    But its been a very old idea, suggested by many to various developers of various games and MMO's, some directly to developer teams (such as myself to Turbine, back when DDO was in early development) with the same responses.
    Quite simply, it would be a huge R&D expenditure with no guaranteed result, since no one in the industry has done it.

    I will say that the first to successfully achieve it though will /win the game of endless quality entertainment, as all current games require a creative human element to them that demands <x> amount of time investment for a static product.
    Will it happen?
    Eventually, yes.
    The technology exists today to support it, its just something that has never been successfully implemented.
    I never said it was a new idea, or My idea. I simply asked if they have considered it. More of a *hint* *hint* sort of way.

    And no it will not happen in the next decade, . . . at least not with that type of attitude/mindset.

    Blizzard did this in a multiplayer format back in Diablo 1. The dungeons all had a standard layout, AND an early form of a procedurally generated map. This was back in the 90's.

    It would further not be about storing every possible map, but creating connection points to small bits of maps. map segment a, b, c and so on and so forth. Then the procedure that runs can connect them in any order it sees possible, never keeping the entire map in the "database."

    As for too much computation time, that could be run on a seperate server that clients are handed off to for the duration of the instance. While it may seem difficult and complicated (impossible for someone like myself to code) I am sure others CAN.





  • Darkstorne
    Darkstorne
    ✭✭✭✭✭
    ✭✭
    OmniDo wrote: »
    Darkstorne wrote: »
    What are you talking about? O.o
    Diablo 3 is the perfect example of why this would be an amazing idea. They handled it incredibly well, and WoW is implementing a similar design for its next expac.
    Diablo 3 is fun to be sure, but the "random" patterns are just the same ol' stuff, with minor variations.
    You dont really see anything "new".
    The goal of procedurally generated content is to accomplish two things:
    1. Eliminate pattern detection (humans are notoriously good at this)
    2. Facilitate near-infinite content based upon finite player decision, while simultaneously accomplishing #1
    Thus far, the only game I know of that has done this is Elite Dangerous, and not because its entirely procedurally random, but rather the developers modeled an entire galaxy within their explore-able space, and thats some 400 billion'ish star systems.
    It would take a single player over 1,095,890,410 years to explore every last unique star system, presuming they could both travel to, and completely catalog one star system every 24 hours. (400 billion / 365)
    Or, it would take 1,095,890,410 players 1 year, however you want to do the math.

    The best one could conceivably do is to model "Tropes" for story content and then a range of "scenarios" for interactive/reactive content.
    Since humans are very good at detecting specific repetitive patterns, only generic patterns (tropes & styles) are regarded as "Fresh" and new.
    Play w0w for any length of time and you have learned all the possible variations for all mob attacks, as well as becoming intimately familiar with the landscapes and having already mapped out the shortest routes from A to B.

    To eliminate this, a game has to have only a "Generic" atmosphere, i.e the area is "Bumpy" or "Desolate", in addition to generic challenges, "The Hounds of scorchland are no stranger to fire", without having specificity that would inevitably become "too familiar".
    With static content players get "bored", the same as they get bored with D3's "random" rifts, eventually.
    You can only maintain the "Monty Hall" campaign for so long before it simply gets old.

    Okay, so you got as far as two words into the OP's thread title and then decided to guess the rest of the topic? He's not asking for an entire procedurally generated game with storylines and scenarios procedurally generated to boot. He just asked if procedural dungeon design like in Diablo was possible in ESO, so that groups could create builds focused on flexibility of strategy rather than knowing exactly what each dungeon would entail from start to finish before even queuing for it. It would be a nice twist on the current formula that's worked well in multiple other games. So I have no idea why the above comment and your original "this has never been done before in gaming history" is coming from, unless you literally didn't read the title let alone the thread...
  • VaranisArano
    VaranisArano
    ✭✭✭✭✭
    ✭✭✭✭✭
    You know, I guess a procedrually generated dungeon wouldn't be too bad if all I wanted to do was fight. If all I wanted to do was experience combat in new scenery, a procedrually generated dungeon would work.

    I just don't think I would enjoy it that much. I like ESO's dungeon stories. Some of them, like Crypt of Hearts, never seem to get old. I enjoy learning the boss mechanics better over multiple runs and finding new and better ways to defeat them. I like honing my tanking and healing abilities, figuring out how to best pull certain difficult mobs together and practicing until my dungeon team can quickly complete a dungeon like a well-oiled machine. I enjoy changing it up by going into a familiar dungeon with random group members so I'll not always doing the same old thing in exactly the same way every time. I enjoy struggling with new content, knowing that if I give up or I have something else to do, I can come back later, try again, and succeed because eventually I will learn and adapt to the mechanics.

    So I guess for me the answer is I wouldn't like it. I don't like ESO combat enough to do it for its own sake in an environment where I can't meaningfully refine my approach and strategy. Other people might like it and that's fine. Its just not for me.

    (However, I must admit to being highly skeptical of ZOS' ability to design sufficient mechanics, bosses, and adds to which MOAR DPS is not the answer to everything in order for these dungeons to actually provide the variety of situations and strategies you seem to want.)
Sign In or Register to comment.