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

Hey everyone!

I'm really surprised that this reached 2mil, and very excited! Recently Sword Coast Legends had a "Head Start" release where their DM Mode was available, and I was admittedly disappointed. Divinity's DM Mode will be well-timed and, hopefully, a great alternative to a limited system like SCL's.

Maybe some devs will post a thread later for ideas (as they did for mods), but it'd be great to brainstorm together here for our DM-mode expectations. n-Space has been facing significant backlash from their community because the community's anticipation did not match the product that was released. If we start dreaming now, I believe we can prevent the same type of calamity from happening here!

So, what would your dream DM-mode look like?

For myself, perhaps the 'picture' I see when imagining a DM mode is taking a total conversion mod (such as creating the Zelda universe) and plopping my players into that world with a great story that I've preplanned.

A few features I'd love:

1) Creating custom items that I either drop in chests, on enemies or...

2) On scripted events. DM-mode should 'play nice' with the editor. It'd be great if a module-designer could create a script function that a DM could apply on the fly.

To explain, let's use Zelda as an example. Link is in the Temple of Time, ready to pick up the sword. I, as the DM, don't want a player to be able to do this until I say so. A mod-designer COULD hard-code that 'x' conditions must be met for this sword to be picked up. BUT, preferably, he/she will implement a function whose conditions a DM sets. EG:



If you were able to implement a way for a modder to do this within the UI, that'd be great. Even with that, though, this form of scripting should still be available so we can get deep and create unique systems! Which brings me to my final point...

3) Extensibility. Even as awesome a DM mode as Larian is surely to create, it can't cover every contingency. It'd be optimal for DM mode itself to be moddable. To add plug-ins to DM mode to allow extended or fine-tuned functionality would cut down on production costs for the Larian team (to a certain degree) AND allow us to make some really great stuff!


Anyone else want to dream with me? What's your ideal DM mode?

Well I hope Larian has an open discussion on what kind of scope would be reasonable before setting things in stone.

http://larian.uservoice.com/forums/...s/suggestions/10001244-dedicated-servers

Agree with all three points, and the second is especially important. A GM needs to be able to override or activate scripts in an intuitive way, and hopefully it isn't gobs of work to manually implement the lines in the editor. Some way to automagically produce the code that lets the GM set flags, change variables, etc.

Also, letting four players control characters with a 5th GMing would be nice, instead of bumping out a player slot for the GM. I can understand if there's technical limitations, though I recall them saying they could run eightish players without too many problems, maybe a tad unstable. It's hard to imagine how four players would work fine, and then five would suddenly break the engine, though of course there will always be a certain threshold to break the engine.

Allright people, like the Mod support topic, I'll gather information on what you would like to see for the GM mode.

Remember that nothing that I will compile here as 100% change to happen but ideas and suggestion, as you all know, are really important to us at Larian.

So feel free to propose cool ideas here!

-- SUGGESTIONS/IDEAS -

- Moddable GM mode
- GM playing monsters
- Player dedicated servers
- GMind the main game
- Add/Remove enemies, traps and such
- Permadeath/new character option
- Multiple GMs
- Dedicated server for persistent modules
- Connection to a database
- Chat channels
- Needs to be 100% stable
- On the fly conversations
- Release "when it's done" over delaying the game with it
- Queue events together or with a timer
- Player become GM and vice-versa
- Procedural generation
- Sessions replay
- Preplan and save an area
- Random Encounter chart for the GM
- Adding personal quest linked to the main campaign
- Hiding player stats from the GM
- A dialogue tool for GM

In order of importance:

1. Be able to DM the main game.
2. Move around the map at will.
3. Be able to take control of any baddie and play along.
4. To be able to add/remove any baddie, trap or loot.
5. Be able to place various NPC types.

I'm not really sure making quests live and/or changing dialog in conversation live really works out.

I know that other game is trying to make the DM a part of the game balance by costing them points to do this or that with a limited pool. Maybe cool, but perhaps not necessary. I guess I would look at it this way, we rather not have a pool and if you play with people you trust probably all good. Random groups... that is where we could see some lousy DM'ing without limits.

I am already skeptical that the GM mode won't become another day/night cycle which gets cut for being a lot more work than anticipated. But I am really skeptical about the idea of being able to "construct levels on the fly".

Even if it's possible, I'm not sure that would be fun to play. D:OS levels and rooms have lots of details and items placed. After you add your new empty room, what will the players be doing while the GM is spending 5-10 minutes adding in all the stuff to it? (Per room!)

I don't think so. At the very least a mode to just control the enemy unites instead would be a pretty fun way to play through the game. Might shake up things quite a bit to quest against a human mind instead of an AI.

I like the idea of the DM playing the monster if he/she wants to.

Adapting to the player's decisions is important so an intuitive system is needed to create situations as the need for them arises.

There should be limits, no one likes fighting an endless horde of enemies every few steps, but at the same time encounters shouldn't be too easy because of a small resource pool.

Permadeath/new character option( My DM does this if we mess around too much.)

Personally I prefer playing the character, but I convinced my DM to buy the game at release, I am really excited about this mode.

I definitely second Stabbey. However, Stabbey, if you ever had a shot at NWN2 toolset, you would have a clear idea of how things could work, and worked amazingly well at that time (9 years ago now ?), both as Permanent worlds and DM directed campaigns, dungeon crawls, one shot stories...
Just another fantasy stretch goal to have people pledge. If it ever makes it in a decent shape (not "the same tool the devs used to create the game, so you have to be a dev to use it properly...", or whatever, you know...), then just make it like NWN2 toolset and system.
But Larian won't make it in a decent and usable state, because to build something like that, you have to think FIRST to create a toolset, and THEN, make something out of it. NOT create something with some tools, and then try to turn theses tools into a toolset, which is what seems to be going on now.

I've preposed multiple GMs in game master mode. Check out my idea, and vote if you like it: https://larian.uservoice.com/forums...248-more-than-one-gm-in-game-master-mode

GM tools definitely shouldn't be limited to curb difficulty for players of the user generated campaign. Why handicap creativity in a co-operative role-playing environment? Fun and fairness can be decided by roleplaying groups just as in a tabletop experience. It's much more important that any features provided in the toolset be allowed to flourish in the hands of a creative gm

Hello everyone, i'm a RP player from the Nwn2 community. I and many friends are very excited by the GM mode and backed the game for it. I will try to explain what we hope for.

Before beginning, a little TD;LR version: The same functionality than NWN2 but more oriented toward persistent modules.

I will try to be concise and only describe the most important system that we hope be available for us and the bonus features, nice but not critical.

- Dedicated server for persistent modules with monitoring tools (number of players).
Bonus: Resources monitoring by scripts, zones ect.. It could be nice to see which script or which zone is killing our performances.

- Connection to a database: For persistence, it should be nice to have the ability to connect to a database. In nwn such system doesn't exist but his essential to persistent modules. Database access are made by reading the server memory for detecting query and writing the server memory for the result. It could be nice to don't have to do that in D:OS2!

- No third party account : Nwn was based on gamespy for log-in into a lobby room where to choose the available servers. And now gamespy is down and we have a tool injecting stuff into the memory for making the lobby work with duck-tape. The most important thing is the continuation of modules, even if the third party managing online stuff is down. Robustness in front of that is better than a server list. Even if there is only a "join by ip" option with a favorite list, there will always be a community maintened list of servers.

- Player character in server side.: For games with friend, it's ok to use a client side character. But on persistent modules, other player may cheat of bypass server rules by using a client side character.

- Security of server side characters: With the assumption of no 3th party account, it must be a way to let a character accessible only to his owner. In nwn2, with the 3th party down, there is a custom script tracking the cd-key of account. If the cd-key is different, the player is kicked. A way to handle that could be the use of a certificate, unique for each installation of the game with high probability, and used by the server to make the right characters available to the player. Server owner must be able to assign a set of character to a different certificate in the case of a lost certificate and player must be able to import their previous certificate to a new installation of the game. Be able to select with which certificate (with different pseudo associated with) the connection to a server is made could be nice too (For dm wanting to keep their "classic player" account separate from their dm account).

- The number of players: Popular servers on nwn have roughly 50 players connected at the same time. 64 seem to be a good maximal size.

- The number of DM: DM should take a player slot but not be limited in number. For torturing many players, we need many DM ^^. DM access must also be restricted by the owner of the server with a DM password for a certificate DM list.

- The party: In persistent server, all players are not friends. So each player should be able to create party with different players in. Other player are neutral by default.
Bonus: Be able to customize the way of setting other players hostile. Ex: a player attacking me is hostile, a player attacking my party is hostile, the entire party of a player attacking my party is hostile, ect... And the ability to set back to neutral an hostile player.

- Chat channels: Server, Map, Local (listenable within a medium range, whisper (small range), Party, DM (thing written by player are only see-able by dm) and private message.

- Custom-ability of the UI and events: I will take in example the death. Some module want a permanent death. This death is only made by DM. When a player dies from a monster or other player, he can have the ability to re-spawn somewhere with a penalty or be taken by an other character to be moved to a safe place for a re-spawn with no penalty.
For doing that, we need to be able to intercept the "rezEvent" or the modify the "UI" to make the re-spawn unavailable if the pj is perma-dead. The moving of the corpse by a character is made by just moving the corpse to the feet of the carrying guy when he move too far after this guy selected the option "carry the target" in a custom UI. When the player select the rez option, the script check if he is in a safe zone and then rez it without penalty or tp it to a safe zone with a penalty. This example is here to illustrate the degree of custom-ability desired.


In nwn the modding community has made an amazing work. the community implemented database access, pet systems, guild systems and guild halls, housing, randomly generated dungeon an many other thing. The only thing we need to move our community on Divinity is just some help to be able to do the same thing (and more!) here without needing too much duck-tape.

With RP friend, we started yesterday to form a team for creating a persistent module. We are our-self programmer (for thing less sexy than games but still programmer ^^) and we were heavily involved in the french rp community of nwn1 and nwn2. So if you need a sort of community created feature list for custom servers (with only the necessary features, the rest can be created by the community), we will be happy to help :'D.

Our RP dreams are now totally in Divinity:OS2 . And we waited such a game for years without finding a potential substitute to NWN2. NWN is a great game but is far too unstable and need to much duck-tape. I really invite you to play a little in a NWN persistent module for finding inspirations.

Edit: Not sure if it's the right thread, you can move it to the other if it's more appropriated.

This feature is important for persistent servers. The world is already created but a DM must be able to make little temporary modification without needing to modify the map in the editor and without restarting the server. You must also consider that players RP between them, XP in automatic dungeon, and do their stuff even if DM are not connected. You are right for "campaign mode" where players are always with the DM.

Per example: Bad guys attacked an house and now stay in. The DM spawn some bad guys in the house when no player are here, moves the placeable to illustrate the fight in the house and then spawn a guy in the city looking for player help. You can do the same by modifying generic maps (inaccessible by the player without DM help) for a specific animation and then creating a portal to this map.

In persistent module, not all DM are allowed to modify the module, but only to create animations for the players on the fly. Only permanent modification to the world are made through the off-line editor.

Not sure but I think they where more thinking of max 4 players and 1 DM. I dont think you want to see this game played by 5+ players. Every turn takes at at least 20 seconds and that is if players play fast. Multipy that by 5 and you already need to wait 2 minutes until a player can make a next move. this will scale up fast and make it boring for a lot of players.

Multiple DM's I'm not a big fan of either. A game can be ruined fast if 2 Dm's disagree with each other.

For the DM I would split up his turn in 2 parts. One is the fase where he can spend action points. All his minions and bosses ofc have a basic amount of action points but it's very low. When he gives them extra action points the minions can do more special attacks. Placing traps, activating special triggers, giving minions a potion this all costs action points and makes sure the DM isn't to powerful.

After this turn the DM can play all the minions that are in combat.

When the DM kills a player he should also get a reward like extra action points for the next turn. Or there could be other ways he could earn actions points like destroying a chest before players reach it.

At the start of the game there should be a limited amount the DM can choose of objects that he can use during the play. Limiting the DM power should make it entertaining for both sides and require a tactical approach.

Maybe I'm thinking this a bit to much in focus of combat but I feel the combat should be limited for the DM.

I'm not sure if I feel for the idea that DM's should be able to add scripts during the game. This feels more like something that should be done before the game starts. I think it would feel like cheating if the DM can do this.

This is an interesting way to see the DM mode. For ourseft, we didn't saw the DM mode as "DM versus Players" but more as the "DM with unlimited powers is telling a story to a group of player". Multiple DM were just here to do different animation with different group of player or managing very big events.

The more problematic in your post is more the turn by turn thing. We are not very familiar with Divinity and i think we didn't even considerate that. If a pseudo real time mode like in nwn2 or pillar of eternity can't be implemented for the multiplayer, a large number of player can't happen. What would happen if a player enter in combat (and switch to turn by turn) and a player in an other map doesn't?

That is not nice and we can't really do something to bypass that without an huge modding of the game engine.

Without the ability to run persistent module, I don't know if it's really worth the time to do a complex DM mode. In nwn the only popular servers were the persistent ones, other with punctual campaigns were very uncommon and are now totally forgotten. Persistent module in an other side are still here and will be here for years until another game with the same ability come up.

Well, we actually know way too little about the GM mode to make a qualified opinion about it.

For once, how powerful is the whole thing meant to be? How much will we be able to "change on the fly"?

Will it be like Sword Coast Legends which means that there are premade dungeons/levels in which you can only place enemies and traps and doors as you like? Or will you also be able to change the whole level design on the fly?

I just hope that GM mode won't delay the main game too much. I'm all in for an easy-to-use and powerful editor and I'm not convinced that the "on the fly" GM stuff is acutally really necessary on top of that. I mean in traditional PnP campaigns the DM already prepared most of the adventure before the whole thing starts as well, which is already possible with a powerful editor for which you don't need programming skills. I don't know whether the ability to change things (whatever these things will be) justifies the additional tool developement and design process.

But maybe the whole thing is more meant as an "easy mode" for the editor?! I mean that as some kind of very -easy-to-use tool that is kind of limited in comparison to the real editor but with which you could do certain things (like predefined enemy placement) very easily. I guess that would be pretty much along the lines for what Bethesda planned for the new Doom (with the big difference of course that there will be a "big and mighty" editor for DOS on top of that). So there will maybe be an "easy mode editor/GM mode" for amateurs and an "expert mode editor" for people who really want to make traditional mods and whole new adventures and stuff.

First and foremost, with the GM Mode, whatever we are allowed to do, it has to be stable. I know Jeff was saying this very thing in that they will have to look at what can be done on the fly safely to keep from crashing the game and ruining people's experience. Traps, enemies, etc... should all be easy to place if they are already created and available as assets to the game. Creating them on the fly while people are in game? Mmm yeah no as you need to test the pieces before letting them hit. Granted, slight modifications to stats wouldn't be too bad.

Effectively I see the setup like this, GM Mode is there to help you bring a more active life to your mod/story. It's how to can react, to an extent, in response to the players. GM Mode is not for big sweeping changes because the players did something you did not expect. If all you want to do is play the random antagonist to the players, this will be great by itself. Especially if you can take control of the mobs themselves. Effectively I see this as a tool to interact with what's already there and spawn in new items/creatures/traps that were either part of the game, or created by you (or someone else) in the mod tools. For dialogue.. mm.. that one is tricky. Setting up dialogue boxes and choices on the fly is not the best really... but I could see allowing GMs and players to 'say' things and it appear in a bubble or even just the info window. This would allow easy on the fly conversations between NPCs and PCs.

However, the Divinity Engine (Mod Tools) is where we build our new areas and assemble our own stories to tell. We lay our foundations and planned encounters here as well as anything that we need customized. Yes it will take time and be some rather involved work. Here I'd really like to see us be given a huge berth on what's possible to manipulate about the game. Assuming the game's logic, skills, and mechanics and such are all using a scripting language this should be fairly simple to allow... basically a "edit at your own risk" type of thing since you'll need to have a good understanding to make such deep mods. While work is going to be done to make this user friendly and easier to use by non-devs, it will still remain that it will take dedicated people to push what can be done to its limits.

I'm fully excited about to possibilities coming by using both tools. However, I will admit, I'd rather see GM Mode not be released with the game as both the enhanced mod tools and GM Mode are going to take a lot of work. I'd much rather see their release come after the game itself rather than delaying the release of the game until the tools were ready as well.

@Kerrida - I completely agree that they should focus on the game and then the tools, but there is also a downside to releasing the tools later. Waiting to long might make it an uphill battle to spread the word. That happened to the Witcher 2's tool, which was released two years after the game's first release on PC. They seem to have more luck with Witcher 3's editor, and I don't even think it contains a level editor.
If Larian release the tools with the game or soon enough that they can still ride the buzz of the release. I think there is a far bigger chance they would become popular. I think it's a strong message to send to the consumers that the GM-mode is there from the start, because it is part of the game and not an after thought.

Oh the release most assuredly should not be well after release of the game. My thinking is basically if it comes down to having to delay the game, or release the GM mode a bit after, I'd rather have the game first. Heck, I'd even be fine with a solid, well polish lower featured set of the GM mode with updates to come to expand on it. I'm super excited for GM Mode though, and it's probably one of my favourite pieces.

I am not sure if this has been stated before because there is a lot to go through on this thread, and I apologize If I am repeating someone else.

One thing that we need is the ability to queue events together or with a timer. This way we can create a poof of magical smoke with some swirls happening at the same time, then a fraction of a second later, we can place NPCs in that smoke (and many other events where you need to place several events or effects into play in a small amount of time.)