Legnds..Northern Realms; juhos msg (rat race) & prj support

[McSkivv]

I'm not sure if downloading on the fly and starting the mod would work in-game. You'd still have to exit the overworld and load the mod yourself. Also, when returning to the overworld, you wouldn't have updated stats, inventory, characters, etc. without starting the mod over. In that case you'd lose your location. Also, what if you wanted a dungeon that started in one location and deposited you in another location upon exiting? You couldn't do that with anything that loads the dungeon separately. On top of that lies the inventory problems.

The other option available to us is much more difficult in nature. Having every dungeon loaded simultaneously in one file would fix all those things, except it would open other cans of worms entirely. First, porting dungeons into the megadungeon would be a pain. Even if you managed to avoid rebuilding the entire thing from scratch, you'd have to go through everything and make sure they all are ported well. For instance, every teleporter would need its exit location adjusted. Also, the file would be huge. On top of that, how much computing power would it take to load all the hundreds of floors involved at once. I believe every floor has to be simultaneously loaded. (Though I don't think that they are rendered and I don't think the CPU pays them too much attention.)

No option before us is perfect. My idea of a website-based map suffers from the lack of awesomeness that a traverseable worldmap has. It also suffers from the problem of inventory tracking. The idea of a massive overworld with the dungeons built in has the potential to be hard to run, hard to update, and hard to download. What we really need is the ability to load sets of floors separately or the ability to save the inventory. With either of these we'd be able to create a stunning experience in the Grimrock engine. One way or another, this project will reach awesome fruition. It simply has too much potential to be left behind in the hypothetical dustbin of scrapped ideas.

[John Smith]

We have most of our options in front of us it seems. If this is going to get anywhere we need to pick what we want to do and how we are going to do it.

I'm in favor of the linear model I proposed. I feel like there is a lot me we can do with a lot less effort and time put in.

[McSkivv]

What we need to do is construct a proof of concept. Let's agree on two potential ways to do this and build a small proof of concept of each. Then we'll be better prepared to make a decision about how we'll do this for the long-haul. I'd like to help with this project if you all will have me, regardless of what method we use.

[John Smith]

I'm not very experienced when it comes to modding for this game, but I'm quick on the uptake and would be willing to help with design. For sure I can write, and if we need that I can cover that base.

[Asteroth]

Wait I already posted this, but don't see it here. Where did that end up?
To repeat:
I don't think of this as a linear thing. Of course I will help whatever is chosen.
Firstly I scream everytime a game takes all my stuff away with questionable excuses(see most sequals).
Second it feels less like a world project when taken as a singe adventure with fixed order.
Can't say I can give a good idea thouugh.

[Arkos]

What we need to do is construct a proof of concept. Let's agree on two potential ways to do this and build a small proof of concept of each. Then we'll be better prepared to make a decision about how we'll do this for the long-haul.

To keep the project aligned as possible to its original goals of being a seamless, traversable, expansive, immersed experience, my vote is for Neikun’s and Jkos’s suggestions of:

Neikun:

I have a solution in mind. Everyone is given a maximum number of floors to work with in their design. (lets say 8) We create all the floors before we start designing. Everyone is given the maximum number of floors to work with. So if we have 10 designers, 80+1 floors. (1 for the hub)
The floors will be named according to designer until release (for easy location)
Example:

Neikun's floor 1
Neikun's floor 2
ect..

Should some people not need the maximum number of floors, their unused floors may be left blank and remain there through the published mod. It has little consequence.
On another hand, those blank levels could be used up by late comers or original designers who want a little more space.

So while multiple .dat files will help break up the downloads, it will also break up the "one complete world" feeling that this project is going for.

And Jkos:

@Neikun, you should use git and github for this kind of project, I think it's the only reasonable way to handle it. It solves both of the problems you stated out. Git is developed for distributed Linux kernel development and version control, and this project is going to be pretty tiny compared to it. There is thousands on programmers working on Linux kernel so git can handle this project easily. Of course it means that everybody must learn to use git, but the basic usage is not that hard.

The idea is that everyone will have their own local clone of the central repository which is located on github, and when you make changes to your local repository you can push your changes to github, and only the changed files are uploaded. Then others can automatically see that you uploaded a new version and can pull your changes to their local repositories (again only changed files are downloaded). Git can handle multiple people working on same script, and if there is a conflict which git can't solve automatically, it will force you to solve it.

Jkos already has a working model of working together with other designers for their community project: grimrock / eob1-waterdeep using github. I think we should do the same for building out the Northern Realms of LoG.

EOB waterdeep severs uses git as a version control tool and github as a central repository. I have contributed to this project and it works really nicely because the actual dungeon file is text based script-file (dungeon.lua). Version control tools doesn't work so well with binary files.

You can browse the project repository here
https://github.com/grimrock/eob1-waterdeep/

It's easy to see what different people has been doing.

The mod “Dungeon Master The Lord of Chaos” http://grimrock.nexusmods.com/mods/149#content is 70+ floors! Neikun’s suggestion of 80 floors+1 (for the hub) I think would be a great start in seeing how a connected world would feel like. If it is done well, then more supporters will come and AH hopefully will have a solution for us to keep it going in seamless fashion. And, if worse case scenario they don’t, then we could go with a hybrid solution combining Neikun’s and the world map solution, but a dungeon file would be 80+ floors like Neikun suggested. Nevertheless, I think starting out with a 81 floor dungeon for a seamless, traversable concept would be the best thing to start with, and while we are waiting for a solution from AH, we could connect that one through some creative means to another 81 floor dungeon (maybe that one would be the Xafi desert and the Toll Islands).

And McSkivv

I'd like to help with this project if you all will have me, regardless of what method we use.

And John Smith

I'm not very experienced when it comes to modding for this game, but I'm quick on the uptake and would be willing to help with design. For sure I can write, and if we need that I can cover that base.

Dude! It would be a privilege to have you guys help! And McSkivv, I quoted you:

One way or another, this project will reach awesome fruition. It simply has too much potential to be left behind in the hypothetical dustbin of scrapped ideas.

on the front page of this thread. This project won’t go anywhere without you guys and others that we hope jump on board!

I have solicited AH for help on this thread viewtopic.php?f=14&t=4639 hopefully they will get back to us soon.

[Neikun]

Players will need to realize that the scope of this project perhaps warrants such a large file size, as gameplay wise, you're getting a lot of content, and a lot of play time.

[McSkivv]

I'm not familiar with github, though I have heard of it before. If I am correct in believing that github would allow us to allow work together seamlessly yet separately, then a combination of Neikun and Jkos's ideas would be best I think. I also think that a website should be dedicated to the mod, and that a world-map type page as I suggested would be a good way to help players understand the scope of the project.

[McSkivv]

May I suggest a plan of action? If so, I would suggest that first we choose a name for the project. Perhaps "The Legend of Grimrock: The Northern Realms"? Secondly, we need a map. We will need a general idea of the overworld, its terrain, and where its dungeons will be located. We'll need to divide it into general level-ranges for dungeons. After that, we need to construct the overworld and its dungeons individually. I would personally recommend creating the overworld and its contents one section at a time. For instance, we might agree that players will start off in X area and then build that area and the surrounding dungeons. We should also decide whether or not we'll have an overarching storyline for players to follow. Also, we should get a list of those people who want to help and what skills they have so that whoever is the leader can assign tasks to the team. Speaking of leader, we should probably nominate a general project leader. I think while we all will have input into decision making, we had better nominate someone to make the final decision in times when not everyone agrees. Is this plan of action acceptable to you all?

Another touchy subject that I feel needs to be touched: the goal of the project has to be changed. Current goals state that we intend to "showcase everyones’ mods in a related endless universe" and that "the amount of dungeons created to be connected to the LoG World is unlimited." This cannot be for two reasons. Firstly, support for loading separate dungeons into the game seamlessly doesn't exist. We have to work within current boundaries, and expand our goals when the boundaries expand around us. Secondly, connecting infinite dungeons to a finite overworld is impossible. The laws of physics and space don't comply.

I humbly recommend that instead our goal be something like this:

The goal of this team is to create an immersive, open-world experience that recreates the Northern Realms as laid out by the developers and as envisioned by the modders and is filled with a variety of dungeons that players can seamlessly proceed through in any order they choose. The world will be developed in two parts: the overworld and the dungeons. The overworld will be a mishmash of plains, forests, towns, and other locations that will link players to the second part, dungeons. Dungeons will be between 2 and 8 floors and will be filled with monsters, loot, and in some cases lore.


This is our world. The developers have already envisioned it; we just have to give their vision life. To do this we need to set an attainable goal, we need to agree to a plan of action, and we will need a general leader who can make the final decision when democracy fails. Last of all, we need a role-call. Who here is willing to commit to the completion of this mod?

Post-Script: All of this is just a suggestion. I don't want to see this project stagnate so I felt that something needed to be said.

[Arkos]

McSkivv,

Great post! Although, instead of one general leader, I think there should be three people (the project chairmen).

My nominations would be:

Neikun, Asteroth, and McSkivv

With people like msyblade, akroma222, Jkos, and mahric as advisors.

Of course everyones’ suggestions and contributions are important! Please don’t be offended if I left someone out or left you out. My nominations are based off of modding experience and interest for this project.

After everyone comes to a consensus for whom the project chairmen should be, then we need one of the chairmen to start a new thread for the project—to make it official and keep the first post of it updated with progress etc. I recommend each chairmen have responsibilities over certain areas of the project.

Chairmen areas of responsibilities:

• One of the chairmen would be in charge of maintaining the new official project thread, the project web site (could create a google site), and github. This person would keep the project organized.

• Another chairmen would be primarily in charge of the architecture and technology of the project. I personally think this leader should be Neikun.

• Another chairmen would be in charge of gathering up resources and help with everything else, supporting the roles of the other two leaders.

EVERYONE should give input for creative direction, but the chairmen are in charge of keeping the project aligned with the project goals and keeping the project rolling.