Larian Studios Forums Divinity: Original Sin 2 Divinity - Original Sin 2 - Game Master - D:OS2 GM Mode - Ideas!

'Should' is a key word.
You have a community familiar with GM. But this idea rather new for many Larian's players. This players were rather dissapointed by GM as KS stretch goal. So I thought, They can be interested in some sort of symbiosis of pnp GM and asymmetric coop-multiplayer.

In fact, I think, Larian will make GM mod like 'developers mod' in Divinity 2. There will be launcher with 2 options. And if you choose GM mod, there will be notifications like: 'Developers are not responsible for failed script, stability etc in this mod'. And you can do anythig you want in game.

I really wasn't much a part of the NWN community, I know what a DM is because I've played pen & paper RPG's for damn near 20 years, it's one of the reasons why I love DivOS due to the freedom of fucking around the one thing that it's missing is irritating the DM because the party has been deliberately ignoring his carefully laid out quest hooks for the past 5 sessions so we can start a mining business or something else absolutely ridiculous.

That's the point of persistent worlds. It's player characters living their own life and doing role-play between them with occasional animations by a DM. So if players want to start a mining business, they can (and many had already do that ;p).

The job of the DM (i'm talking in the case of PW, not in general) is to give to the player "tools" to help interactions between the players, make the world living, and some-time creating major plot-line within the module.

It's not just about 50 players fighting monsters all the day. There are quests with any fight involved, political intrigues and factions play.

In NwN, some of these worlds live since ten years, with player connected several times by week. But now, NwN start to becoming really old and a little wearying, so you can understand the hype in the NwN community with the DM mode of D:OS2 :'D

For persistent worlds to really be meaningful in D:OS, you would need respawning enemies and encounters, and you'd also need someone constantly maintaining the world and ensuring a constant flow of new content. Without that constant maintenance, your players would run out of things to do.

That is theoretically possible to do, but I think the biggest obstacle would be finding some kind of DM who has enough time to handle said constant maintenance in a game which, unlike Skyrim, does not have built-in systems to do that automatically.

We are already a team of developers with a long experience of PW in NwN ready to do the same in Divinity:OS2 if the DM mode and the editor has enough possibilities.

The respawn mechanism can easily be implemented in any game allowing to spawn dynamically a character. In fact, a PW need the same system than a MMO, with dungeons, quests... An even more.

To the fans of persistent worlds: ask yourself 2 questions.
- Do you have a server that is permanently on and that can host many players? Larian said they do not have servers. Do you want to check all the time what the players are doing on your server?

- Have you ever played a game with 5 or more players at the same time and with turn based combat (one creature acts after the other)? In all games with pw I know (not so many, I admit) all players can do something all the time, they do not have to wait until all other players finished their turn.

I think GM mode will be one of two possibilities (maybe both):
- some kind of asymetric pvp
- simulating a PnP game (up to 4 players play a campaign that is designed and moderated by the GM)

I do not say that persistent worlds are impossible. But it will be lots of work to get it working and the final result would be very different from the D:OS we know.

Yes, the dedicated server is require and we already had it (like many PW). Secondly, players can (and are supposed) to do their own stuff without a DM behind them. In a PW, the majority of the interactions are between the players, in many PW players could play for weeks without DM. In practice, several DM are connected every day for 3 or 4 hours.

The best way to imagine a RP PW is from the point of view of a MMORPG. In a MMORPG, you don't need DM to do explore the world, do dungeon or other stuff like that. If you want you can RP with other player and some players do such RP for hours everyday.

In a RP PW, the main goal is to do RP stuff and there is zero Out Of Character interations. But good PW also have MMO and SandBox system, like dungeons, resources collection. So when a DM is not here (mainly for interactions with non-playable characters), the player can do these MMO stuff.



I heared somewhere that Divinity accept local turn by turn, i.e. players away from the combat are not in turn by turn. The miss-conception with PW is that player are fighting all together. In PW, players are not friends or even allies, they can be enemy. Usually there is a main HUB, like a city or something else where players can meet. In these city, players don't fight and mainly talk (and even, in PW 80% of time, it's just talking). So the TbT is not a problem. When players fight (or even are with a DM), they are usually in little groups (it's why we need several DM).

Fight actually involving more than 10 players are quite rare and anyway just a mess. Instead of a big fight with wave of trash-mobs, i think it can be managed with lesser and harder mob, taking advantage of the turn by turn setting. In a PW, there is no reload the past savegame or other stuff like that and if you are killed during an animation or by another player, your character is dead, like really really dead and you have just to create another. So I think that having a Turn by Turn setting could actually be quite welcomed. It's never nice to die because an assassin killed your 2years character in 10 sec while you where looking elsewhere :p.


It's a little like i am repeating myself by we don't ask (or don't even want) that Larian create a "MMO DM mode" for Divinity. We just want the tools to create our PW, more precisely:
- Dedicated server
- No hard limit on number of players
- Flexible party system (different players in different party, with no party by default unless the player join one)
- Flexible DM system

For crafting the world, the MMO/sandbox system and all the other stuff, we can do that ourself.

I think the GM mode will be simulating pnp campaigns like no game has ever allowed before. It will be the real successor to the Aurora toolset.

When I see how DOS came from after it started as early access and now what they have done with the enhanced edition, I have no doubt it will be awesome.

The pre-alpha build they showed during the kickstarter campaign is gorgeous and with 15 months of more production ahead, nothing is impossible, especially with such an epic kickstarter final!

You heard? As in, you haven't played D:OS 1 yourself? I mean, what you heard is true, but if you haven't actually played D:OS yourself, you might be carrying in some assumptions about the game which can lead you to make some incorrect assumptions.

Without modding, D:OS has no day/night cycles, NPC schedules, no enemy respawning, no resource respawning. I'm not sure that there will be enough things to do in the world to keep player interest without pretty frequent content injections by the GM.

The hard limit on players is an engine limitation, I believe. I don't think it can handle more than maybe eight or so players at the absolute most, and there is only one party. You don't have to stay with the party, but all PC's portraits will be on the screen, so there isn't really such a thing as two different parties, never mind enough players to require more than one GM.

There are resurrection scrolls in D:OS so a dead player can be brought back.

If you haven't played D:OS, you really should play it to get an idea of what it's like to adjust your expectations.

As a newbie to the forums, I'll introduce myself as someone who is a massive fan of D:OS and a backer of D:OS2... I hope you like my suggestions.

I really see the GM mode -- as many others here have described it -- as an enhanced version of P&P RPGS like D&D.

I'm personally not very interested with the idea of persistent worlds, or deep modding of monsters, skills (ie custom skills), etc. I feel that these are pretty cool ideas, but entirely unrelated to the idea of a "GM mode". That said, it'd be awesome to be able to include any modded items, skills, NPCs, creatures, etc that are created by the modding community while in GM mode. However, for the most part I think the mechanics built into D:OS2 will be plenty versatile.

I feel the key focus of the GM mode is to be able to create a story in as close to real-time as possible. I like the idea of controlling enemies in a battle, but for the most part I'd prefer to trust the AI and move elsewhere to ensure the next part is prepared. I'm sure the AI for D:OS2 will be be pretty cool.

Essentially, I feel that GMs should be able to drop-in and drop-out of combat -- either letting the AI control it or not. By the same token, I feel that the GM should be able to drop-in and drop-out of NPC dialogue. Even if a dialogue has been pre-made, he should be able to intervene at any stage. Also, maybe half way through combat one of the enemies wants to beg for his life, or offer the player a chance to surrender?

Anyway, the whole point of having a GM in a P&P game is that a GM is able to react intelligently and dynamically to the players. This is different from just playing a board game where the rules are clearly set, and everyone is restricted enough so that their actions all have clear and explicit effects. In P&P, it's hard to predict EVERY single possible scenario beforehand, but with enough preparation and creativity it's possible to react quickly.

We already have the suggestion about Procedural Generation. I couldn't find this explained in this thread, but my understanding is that this is a tool to help with quickly building maps in a pseudorandom way. In my experience, some of the Heroes of Might & Magic games had good "random map" generators, where you could set a few conditions, and it would generate a playable map. I wouldn't have to agonise over every little detail (am I giving this player too much stuff?), and so I could invest more time on the "big picture" of designing what I wanted to design. It would be awesome to have the ability to use this kind of tool to generate entire maps, maybe making a forest, and then decide to place a generated town of a friendly faction, maybe place a town of an unfriendly faction (eg orcs) somewhere else -- with a suite of minor NPCs already included. Hopefully, I could choose the size of the town, generate a building to drop in somewhere, and then add a few NPCs.

This actually leads into another idea I have:
Territories and Factions As part of planning the story, I would need to think about the politics. An example of factions in D:OS is the orcs (faction) who are fighting with Cyseal (faction). Immaculates would be yet another faction.

But factions wouldn't only be limited to only cover sentient creatures. For example, if there is a scary forest full of giant spiders, you could create a faction of giant spiders, set the kinds of creatures that fall within that faction, and then zone the forest as the territory of faction of giant spiders, and can generate spider encounters within that territory.

The factions would help to determine how the AI would react when an a Cyseal soldier meets an orc, or when an orc meets a giant spider.

So I would like to be able to design factions and map out territories. Ideally, I would be able to design what kinds of NPCs/creatures typically fit within that faction, what kinds of classes and equipment they would tend to have, what kind of "towns" they inhabit (eg the style of a human town is very different from a bandit camp which is very different from an orc town) and how they would react to other factions.

This is the main kind of modding I'd be keen for in a GM mode.

Further to the earlier idea that the GM should have freedom of movement, and be able to let the AI handle itself sometimes, I also feel that we should be able to keep track of the players while we're distracted on other things. We wouldn't want the entire party to be wiped out simply because we're not constantly watching them, or for them to do something stupid like attacking a guard in a town, where suddenly the AI doesn't know how to react. There should be some kind of Player Tracking. Possibly some kind of a split screen?

Finally, the GM needs to be able to quickly Pause the Entire Game -- even if the players can't -- for those times when he really needs a minute to get things ready.

What I expect from this GM mode is the possibility to use it even as a solo player, because some people are solo players and still they should have a chance to use the function of the final stretch goal. Having the opportunity to set up a whole dungeon filled with packs of monsters and being able to play through it afterward.

I think your hopes are set too high, Ayvah.

A video game GM by definition will never have the freedom and flexibility available to a pen and paper GM, not with today's technology anyway.

Someone might have made the suggestion of procedural generation, but I don't believe that Larian has the tools or resources (time and money) that can make the code which will create a procedural generated D:OS map - at least not one which has a sensible layout and interesting features. Procedural generation is difficult to get right, and if they wanted such generation to produce maps with the kind of quality that goes into the hand-crafted maps, they'd have to murder the rest of the game in order to get the funds and time.

Factions already exist in the game at least.

I certainly do not expect GM mode to be anywhere near as flexible as a real P&P game.

However, I feel the most important part of the experience of having a GM is the flexibility of the storytelling. Flexibility of actions is less important. That is, I feel it's not important that GMs are able to improvise a mechanic like climbing a wall or jumping through a window. But if it's appropriate for a character to react in a certain way during a conversation, then the role of the GM would ideally be to facilitate that conversation, and enable the story that unfolds.

Procedural generation is something above and beyond that, however, and it acknowledges that GMs are likely to be very much amateurs. I expect we would have a ratio of 1 GM for up to 4 players, so potential GMs (like myself) lack the talent or skill we might expect from "modders" who are releasing their mods to much larger audiences.

I'll leave it up to the developers to work out what is practical from the perspective of budget and development time. Procedural generation doesn't have to live up to the expectation of being as good as handcrafted maps. Even if it only gets you halfway there, then that's half of the work done already. The ideal is to minimise the amount of micromanagement a GM has to do to get something functional, while giving them as much control over the results as they need.

There are modest versions of procedural generation. One is improving the terrain "painting". For example, I could paint a forest of trees over a particular area, and the engine would select individual (and varied) trees to place in that zone, instead of forcing me to individually select and place every single tree. The game could generate suggested names for NPCs, stock for merchants, etc. The game already has random loot, and I expect we'll be able to generate random encounters. These are all small parts of this picture that I don't always want to micromanage.

Procedural generation is something that works best when it's part of the design from the beginning, because if the tool is good, then Larian would be able to use it themselves when making the game. Procedural generation is already part of the game. I just really want to emphasise my view that "more is better".

I do feel that procedural generation has a lot of potential to be a key component of the GM experience.

I think GM mode is inherently a multiplayer mode.

However, I think that the sliding scale of expertise for using modding tools looks like this:
Professionals/Developers > Modders > GMs.

I think GMs will need certain improvements from the modding tool that will help you to get the result you want, even though I imagine it is outside of the scope of the GM mode itself.

It isn't that you couldn't do a dungeon yourself after modifying it in GM mode (afterall, sometimes you will want to test a dungeon)...

It is that making things for you to try out in single player is better left to the normal editor.

GM Mode is more about the ability to change and control things on the fly during an actual game.

I'd say the point of a thread like this is to dream big. While those ideas might not be implemented, if we don't dream big than the developers may not know what we MOST want. Even if those are impossible creations, they can use that to guide the mode's creation.

I like the procedural generation model Ayvah has put forward. It's still a far reach, surely, but if GM mode is to be accessible AND creative, custom and easy map creation would be a huge plus. I'd rather other resources take priority, surely, but if it was something they had time for... great.

Here's a competing game which has a DM feature, most will play it real-time and people will complain it is too shallow. However this link is a group of players and DM taking their time. So what is looked at as shallow, gains depth based on style of play. Not necessarily my style, but if players have patience a DM can add a lot.

Start at 15:00 mark.

http://www.twitch.tv/jakonian/v/21900929

I preordered this game and then after playing all the headstarts, I asked for a refund. I like creating scenarios but this game lacks all the basic tools you would need to do it properly.

So I do hope Larian is gonna make it much better. And I know they will actually.

Love me my D:OS, but I can't say that with certainty. What are some good meaty modules made with their mod tools for D:OS? It never really happened. I think one can argue while SCL's tools have more limits, it will be easier to create content and share it. I think you'll also see some patches to help with the basics that are needed. D:OS mod tools aren't in development that I know of, that is done, it is what it is and that didn't happen.

This is a little hard hitting but fair. I sort of feel DM/GM is a trap in which I hope they are accurate in articulating what it will and won't do. Fans will have wild expectations that most likely won't happen. I'd say tread conservatively.

Watching some streams of SCL, I still like the idea of the DM taking an aggressive playing role, it is more natural for video game flow of getting all involved with some action. However the above link shows that if everyone is patient, you can have a role playing experience, it is slow and imo a very limited audience, albeit a loud audience.

There's nothing wrong with dreaming big. But my concern is that Larian's reach often exceeds its grasp, and the final product inevitably falls short.

That might be a problem with all game development in general. But those other games are developed behind closed doors, whereas Kickstarter puts the lofty ideas of developers up front with stretch goals and makes promises that if they get X dollars, they can definitely put feature X in. And Kickstarter is so new that people haven't developed the necessary jade-coloured goggles to make them temper their expectations.

So no, I will NOT stop suggesting to people that their ideas might be too ambitious, and that they should temper their expectations to something vaguely realistic.