OpenMech - A community driven design for a flexible Mech game.

[SpcWest]

So what do we need to get started on the design portion?

Title: OpenMech

Objective: An open sourced framework & editor for developing BattleTech/MechWarrior games, of numerous styles.

[cmunsta]

Hey folks, sorry for being gone for a while.

[member=2511]Legend[/member]
Hiya, things are as good as can be expected though rather busy with things and simply can't dedicate too much time for other things (hence the late repy to this).

[member=4225]SpcWest[/member]
At this point we need to come up with all the things that the engine would need to handle and how it is to handle them. The first post doesn't even come close to scratching the surface really.

As a quick example of what I mean, the engine will need to handle any graphic shaders to be implemented. But how will the engine specifically used such things. Further, do we want to simply implement a set of stock shaders or allow custom shaders. How would these be implemented, loaded, etc.

We need to start at a high overview, then break it down to finer details. Not everything needs to be decided on in advance, but enough detail needs to be given so that those implementing things have a good idea of what to do and where it is going.

[PLEXI]

This project idea has certainly piqued my interest cmunsta.

However, I do have some concerns about the proposed scope of the project and it's overall viability. Focusing on an RTS game emulating the MC games is far more ideal.. But if you are truly set on creating such an open ended program; my suggestions:

The current proposed scope is far too broad to achieve any of the goals set before it realistically. I recommend refining the core game-play mechanics down to those of a a hybridized RTS / FPS. Allow for these difficult core mechanics to be developed side by side and then create a partition that would allow users to choose either one separately or perhaps as a hybrid. Alternative 'modules' such as the ones focusing on CBT or Side scrolling, should be easier to develop once a foundation is created.

• Unity 5.  Great for rapid prototyping and very flexible overall for most users. 
• 3D graphics. Leave 2D for things like billboards; particles and GUI elements / etc. This will give you the flexibility to switch from RTS to FPS while keeping a consistent level of quality and keeping the asset list as small as possible.
• Editor. (Main game should take focus first) A lot of this will depend on how files are stored and retrieved for use in the game. This should be a separate program made available for download later. It needlessly increases the file size for those who may not want to use the Editor.

I look forward to following this projects development. Please do not hesitate to notify me if there is anyway I can help contribute.

[Deathshade]

I think this is a great idea and I would like throw in my hat to help with this. I know you guys are just beginning in this so decide on you coding platform first so I can brush-up on the new code. 

I think getting the core conversion done for the MC code could be a good focus. With the core being the UI, maps, mechs, objects and maybe even the AI that drove the campaign side of MC.

[SpcWest]

• Editor. (Main game should take focus first) A lot of this will depend on how files are stored and retrieved for use in the game. This should be a separate program made available for download later. It needlessly increases the file size for those who may not want to use the Editor.

Thanks for the input! I liked the other points presented, but this one stuck out to me in particular. I don't know that there's an actual game proposed or conceptualized yet, which is likely related to the openness of this proposal. Maybe we should start pitching games to focus on as a driving point for further development? Are we all really just thinking MechCommander 3?


EDIT: Just had the thought, what if we do a game called "Nova Commander"? I'm thinking first/third person tactical simulator, giving the player capability to either directly control their mixed unit of BattleMechs & combat or support vehicles from:
- entirely their cockpit
- entirely from the C&C vehicle using top/down or third person maps
- as a mixture of both


Using a Nova as opposed to an IS Company or a Clan Star would permit mixed units of vehicles, aerospace and battlemechs, while also offering up a pretty diverse selection of lore to base the game off of. You could even explore the Clan Wolverine annihilation if you wanted.

From a technical stand point, this would allow us to focus on a specific game as our goal rather than an abstract engine. Additionally, the diversity of game play style would allow one to use the framework and editors produced by this effort to make a variety of new titles, ranging from more tactic focused MechCommander/Battletech style games to more simulator focused MechWarrior style games. I believe even aerospace fighter scroll shooters (think Galaxian) should be possible.

[delMar]

I'd love to see this come alive. First thing I'd do would be to re-create The Crescent Hawks' Revenge :-)


As for the game engine, I love what's able with Unity.
For an undertaking like this I'm afraid it's the wrong choice.
While Unity shines at implementing games, it's not so well suited for creating moddable game creation kits.
When you make things editable in a Unity game, you lose much of Unity's built-in power because it prefers to read resources from unity-style resource packages.
Shadowrun Returns is built with Unity and it comes with an editor. While it can do lots of things, it doesn't run stable and the game is still not fully moddable because of the reasons I just mentioned.


I googled a little bit and I'd suggest using the Spring RTS engine.
https://springrts.com/


I did not work with it myself, but I think it provides everything that's needed for a rendering engine and it's quite up to date (last commit 3 days ago).  It's really a RTS engine, so it should come with most of the very basics that a project like OpenMech requires.


Plus, it provides scripting with LUA.
As a couple of game projects have already been created, I'm pretty sure that some editing appliance is already available as well.


OpenMech would still require custom code for the briefings, mech bay, etc.
Once the engine is done (no matter if Unity or not), the real work starts with generating content like mechs and other models/sprites, scripts, a story, etc.


So, for speeding up creation of any OpenMech campaigns, some common, basic resource set with mech models, buildings, etc would be great.


Just my two cents, please feel free to ignore.

UPDATE: someone started working on MechCommander for SpringRTS already http://sourceforge.net/p/mwspring/code/HEAD/tree/

UPDATE 2: here's the more recent version:
https://github.com/SpringMCLegacy/SpringMCLegacy

[SeanLang]

[member=3325]cmunsta[/member] bump! I'd love for the community to take off with this! Any engineer, artist, designers out there?!

[cmunsta]

Unfortunately, I have even less time on my hands these days to do much of anything. As an example, I'll be pulling a 14 hour shift at work this Friday.

Further, as it doesn't look like too many people are interested in getting involved (based on responses to this thread) and there are other less ambitious projects proposed (based on the sub-reddit and other posts here, though I don't know if any of those are moving forward), I'd have to say it is a no-go on this idea.

If anyone does want to go forward with this, or even just mutate this into something else, go for it. I can post a few ideas and such but I really have a lot on my plate and Mech stuff isn't very high priority for me.