[Kagato]

Citadel: A massively multiplayer platform game concept

Latest info is always at the project's brainstorming page: http://citadel.wikispaces.com

--------------------------------
Summary:

The game is a side-scrolling platform-based RPG/action game, in the style of Castlevania and Cave Story.

The game takes place entirely within the Citadel, a vast castle-like structure. No one knows what lies outside its walls. Long ago in the forgotten past, the Citadel suffered some form of mystical catastrophe, and the world was fractured.

The castle exists in a realm of perpetual Night, pristine but sterile; no plants grow there. The people make what lives they can, enjoying a magically-imposed peace (no human can harm another), but always on the defensive against the ever encroaching monsters from above and below.

However, there is a ritual that allows a person to break through the barrier to the realm of Day. In daylight, the castle is in ruins, and overgrown with wild plants, but no other creatures disturb the silence -- only the humans who dare intrude. Many of the plants are deadly, and there is no spell of protection preventing attack from an enemy. Still, there are riches to be found there; both among the Trees, and from the bodies of fellow travellers.

Some small areas of the Citadel exist between the realms. The protection spell holds, and no monsters can enter, so these halls provide welcome sanctuary for weary adventurers. Perhaps they represent the original nature of the Citadel. But the Twilight Rooms are few, and their numbers may be dwindling...

--------------------------------

If the concept sounds interesting, please take a look at the project's brainstorming page. There is much more detail there, and it is updated regularly as the concept evolves.

I'm looking for some feedback and information on a number of topics:

* Thoughts on the game concept itself, whether you see any gameplay issues or not, further ideas...

* Suggestions for a game engine, platform or supporting libraries. The less that needs to be written from scratch, the better.
Note that the game will be optimised for the Pandora, but it won't be exclusive to it. (Otherwise you'd end up with an MMO with no players...)
Ideally the software will support the Pandora, Linux & Windows; possibly PSP.

* Assistance with development of the project.
I can manage production of the initial art assets. (later on, having a few extra people would speed things up)
I could probably write the basic platform engine okay too, but I have no experience with developing multi-client network support.

Perhaps there is even an existing game that could be extended with the required features.
Interested developers, let me know.

This post has been edited by Kagato: 24 October 2008 - 04:28 AM

[sindbad]

In no particular order:

I imagine certain areas could get very crowded and in 2D that makes stuff very hard to tell apart. Do you have any plans to minimise/control this?

Pixel art is unacceptable these days, you must use SVG if you want it to look good everywhere and at any angle.

For an engine, http://cocos2d.org/ is pretty cool (python only), I haven't tried others. There isn't much point in using a 3D engine, but it would be possible.
You could use a 2D physics engine to make stuff more fun and easier to program and here you basically have 2 choices: box2d or chipmunk. After using both for a bit, I found chipmunk to be more general, customisable and robust. It's also written in C, which makes using it from other languages that much easier. Personal taste has a lot to do with this.
In a worst case scenario, you could use SDL, but it's 2D graphics is a bit slow and it's certainly not an engine.

I would suggest starting it in python (or the other popular dynamic language of your choice) at least for the prototyping phase. Incredibly short turn-around times. However, you must make sure the library(ies) you use in python are availabe in C or D or whatever else you want to use later on.
You could keep it in python and only write the slow parts in C and package them as python modules. This has worked marvelously for me in the past.

I'm willing to help, but that depends on your choice of language, libraries and architecture and my free time (uni comes first).

This post has been edited by sindbad: 23 October 2008 - 10:00 AM

[Creature XL]

WOW!

Just read the linked page. This absolutly sounds like agame I always wanted to play. I like Castlevania much. If it is turned multiplayer with all the cool things there are in "standard" MMORPGS ( I played Everquest for some months) its da shit, dude! I really like it.

However, I must say I don't believe it could be done If the project takes off I most probably join ( I am programmer).

This post has been edited by Creature XL: 27 October 2008 - 10:48 AM

[Milamber777]

Very excited about this concept. I have experience in class balance, mechanics (basically, "item balance"), and testing (when it gets to that point) and will help as best I can. (Both balance & mechanics experience from MMOS). No current programming experience to speak of though. Good luck with this, and I look forward to seeing it progress!

[Yamara]

I imagine certain areas could get very crowded and in 2D that makes stuff very hard to tell apart. Do you have any plans to minimise/control this?

One possibility is to limit the number of people who can join an area, similar to how WoW has instances where only team members join an area. You could then use a format like Golden Axe for the common areas, where you have a bit more 3D. This would allow you to use the same art, but also allow more space to see different player sprites.

I'm not familiar with many games that make good use of multiple players in a platformer. I recently read good things about the old Chip 'n Dale NES (SNES?) games, in that regard, allowing you to pick up other players and throw them. Things like that could make for more interesting puzzles and interaction.

[Mattz]

I read through your ideas, and so far it sounds amazing! I would love to help with the programming, but I have no decent amount of skill with using C and/or C++. ( I know HOW to program a game, but I'm not strong enough with C++ to use my knowledge to create a graphical game.)

Very good concept for a game, and I will haelp if I can.

[dentrado]

[...]

I'm not familiar with many games that make good use of multiple players in a platformer. I recently read good things about the old Chip 'n Dale NES (SNES?) games, in that regard, allowing you to pick up other players and throw them. Things like that could make for more interesting puzzles and interaction.

I came to think about lost vikings (for snes I think) which is a puzzle/sidescrolling game where you control three vikings, each with different abilities. The goal is to get all three to the end of each level.

[PoisonedV]

Sounds more attractive as a networked RPG than an MMORPG.
that, or several small servers, with limits for players depending on the amount. use python, its portable and has alot of great libraries, like pygame, pyglet, and library warppers for box2d and chipmunk, and some nice engines, too, like cocos.

(cocos for iphone is opengles http://code.google.c...ocos2d-iphone/)

also, svgs for 2d game = ugh

I especially like the seeds idea. in my opinion, MMORPG's need to move away from being 'MMORPG'S and more towards being online RPG's. they need more interactivity, chances to actually shape the game world. player housing, player inhabited governments, player ownable businesses and stuff etc like in EVE and Puzzle Pirates are progressive from the current state of most MMO's.
also, the time should be based on ticks so that server owners can set whatever speed they want...

This post has been edited by PoisonedV: 23 October 2008 - 09:59 PM

[Kagato]

I imagine certain areas could get very crowded and in 2D that makes stuff very hard to tell apart. Do you have any plans to minimise/control this?

Honestly, the game is unlikely to get sufficient players for it to be a problem; but I agree that handling for overcrowding should be designed in from the start. One possibility is to divide the map up into rooms, and "lock the doors" once the maximum number of players are present in a room... but that could make navigating the map kind of difficult.

The server should be able to keep track of thousands of characters, and the client screen will only big enough to render dozens of players at a time anyway, so I'm hoping technically it won't be a problem. Visually it could get very cluttered though, so big popular areas will need to be fairly vertical in their design to provide more standing room. Imagine the main player halls having lots of stacked levels that you could jump up or down to, sort of like this:

Pixel art is unacceptable these days, you must use SVG if you want it to look good [...]
For an engine, http://cocos2d.org/ is pretty cool (python only), I haven't tried others. There isn't much point in using a 3D engine, but it would be possible.

Cocos2D looks pretty good, and as PoisonedV points out it has an OpenGL-ES port for iPhone. I'm not sure how many of its shortcut functions could be used, but they might be handy for animation tweening. However, it doesn't seem to support SVG (and neither do the other python game libraries from what I can tell), so even if I design the sprites in SVG they'll probably have to be rasterised anyway.

You could use a 2D physics engine to make stuff more fun and easier to program

Physics would be cool, but I'm sure synchronising simulations over the network gets pretty hairy. There's not a great need for it in the concept at this point; simple platform game physics (ie pretty much just gravity and box collision) should be sufficient.

I'm willing to help, but that depends on your choice of language, libraries and architecture and my free time (uni comes first).

That's okay, I don't have much free time either; this won't be a super-fast moving project

Just read the linked page. This absolutly sounds like agame I always wanted to play. [...] However, I must say I don't believe it could be done

Glad you like it.
I don't see anything inherently impossible in the requirements though; the biggest hurdle will be the network library.

Very excited about this concept. I have experience in class balance, mechanics (basically, "item balance"), and testing (when it gets to that point) and will help as best I can.

Thanks for the offer; when (if) it gets that far, I'll get in touch.

I'm not familiar with many games that make good use of multiple players in a platformer. I recently read good things about the old Chip 'n Dale NES (SNES?) games, in that regard, allowing you to pick up other players and throw them. Things like that could make for more interesting puzzles and interaction.

There really aren't many multiplayer platformers, which is why it seemed like an interesting design. (Of course, there may not be any because it's a bad idea, but I guess we'll find out.) One example of a (small) number of players in a platformer is Super Smash Bros, for the Day (PvP) side.

Players throwing each other is certainly a possibility.

Sounds more attractive as a networked RPG than an MMORPG.

I'm sure it'll start out that way during development... and if it can't be made to scale up it might stay that way.

I especially like the seeds idea. in my opinion, MMORPG's need to move away from being 'MMORPG'S and more towards being online RPG's. they need more interactivity, chances to actually shape the game world.

I haven't planned out all the possibilities that planting Seeds could provide, but I'm sure there's lots of room for innovation there. (Planting your own house is certainly one...)

[PoisonedV]

There are pyglet sub-libraries for svg's, which might be easy to integrate with cocos2d (it's open source, after all)

[Eniko]

Pixel art is unacceptable these days, you must use SVG if you want it to look good everywhere and at any angle.

As an accomplished pixel artist I would like you to substantiate this claim somehow.

[sindbad]

Pixel art is unacceptable these days, you must use SVG if you want it to look good everywhere and at any angle.

As an accomplished pixel artist I would like you to substantiate this claim somehow.

It's partly a problem of manpower, because you can't really make enough versions of each sprite for each different resolution and aspect ratio combination you may come across. Vector graphics clearly has the upper hand here. Look at KDE4 and how well icons work in it.

The other side of the problem is rotation. SVGs can simply be regenerated either beforehand or on the fly, depending on hardware capabilities and needs.

Even SVG starts having issues when you need Z-axis rotation. Using 3D models rendered in different positions is a partial solution, but you usually want full 3D at this point.

Pixel art is pretty cool and I meant no disrespect, but it just isn't really useful anymore.

This post has been edited by sindbad: 24 October 2008 - 08:53 PM

[PoisonedV]

so you get jagged edges- big deal. they still look great to me.
if you could, however, do something like flash on low quality- you get the raw pixels, but scalable?

[Kagato]

As long as your sprite library does decent alpha and bilinear filtering, you shouldn't see a noticeable difference between a rotated vector sprite, and a double-sized bitmap sprite rotated and scaled by half.

Vector sprites are more elegant of course (and have other advantages, eg. colour replacement), but if the bitmap libraries are faster and better supported, it's certainly the simpler option.

[PoisonedV]

gameplay wise, how would items work? would monsters drop items and gold, or maybe some alternate form of currency taht could be traded in? it always seemed illogical to me that monsters had hundreds of GP on them at all times. what sort of stats would there be? would you have one that strictly affected things like magic and combat power, or ones that gave you physical legs-up, too, like higher jumping, etc? would there be player owned stores?