Turn based RPG?

[pixel_legolas]

I just wanted to know how hard it would be to make turn based rpg. Have you gekido done any experiments? I think I remember that you have said something about it?

I am very eager to know more and would like to make my first major project as turn-based RPG.

Please guys, any info would be nice

[zknack]

Pixel,

This has been a seriously tempting one for me as well; I think it's do-able through some serious
edits of the chess game, most all of my assets I've been making are fantasy rpg-oriented (with
the exception of some that are fairly universal).

Are you thinking of a free-move 'world', then turn-based combat more akin to a final-fantasy
or final fantasy tactics type system; or are you thinking more along the lines of something that
is entirely turn based?

My latest tests have been for massive 'world maps'.

I'm interested to know where you're heading with this one.

Let me know what exactly you're planning, and I'll help how I can.

[pixel_legolas]

hm, basically I am just in the face of "I want to try something that has RPG written all over it". What exactly I want I don't even know myself but I have played all FF and love em.

I think I want free roaming worlds with random battles that just pop up. It could be that you actually see the enemy and when entering a zone the battle system/UI starts with timers and all.

I am very interested on what we could do with this because I feel I need a sharp project soon

[zknack]

For random encounters I've been planning on using 'event proximity' triggers scattered through the levels, sometimes with some overlap- that way you can have multiple 'likelihood of being attacked' sections all on the same map.

Combat maps as well as enemy types could be pulled by the event trigger from the map that the triggers are sitting on- that way it would be simplifying the codebase.

A basic step by step of how I'd break down combat:

1: Player's moving on map section
2: Player enters invisible proximity event trigger for combat
3: Event trigger checks percentage, if it keys combat continue to 4, if not see 1.
4: Trigger does following:
       a.) Trigger checks map attributes list
       b.) Based on map attributes list, sends player to combat map
       c.) Based on map attributes list, loads enemy type
       d.) Checks player list- Based on player list begins to load appropriate party members
       e.) Initiates combat GUI
       f.) Goes to step 5
5: Combat Is initiated & does following
      a.) Shows appropriate party members that were loaded
      b.) Loads any 'combat entry' animations for party members
      c.) Shows appropriate enemy models
      d.) Loads any 'combat entry' animations for enemys if available
      e.) Kicks off party & monster idle combat animations
      f.) Player makes selections on GUI
      g.) GUI Selections key off actions and appropriate player animations
      h.) Background work is done to establish effects
      i.) Player animations key off response animations and/or actions from monsters or other party members
      j.) Monsters execute actions
      k.) Background work is done to establish effects
      l.) Monster animations key off response animations and/or actions from Players or other monsters
      m.) F-L Continue until battle is resolved
      n.) Cue Step 6
6: Battle is resolved!
     a.) Load appropriate victory animations & music
     b.) Load appropriate screens for victory or defeat
     c.) Show battle recap
     d.) Give awards or penalties based off of monster type
     e.) Items/rewards are added to a inventory / item list screen
      f.) Teleport player to correct map location to repeat from step 1.

Your thoughts?

[pixel_legolas]

Well that basically sounds like a perfect old turn-based system to me. So let's see what this can lead to

[zknack]

Well, if you want this to turn into a coop'd project, how in-depth (or basic) should the stats be?

I'd be willing to put my projects on hold for this :-)

Should we allow player assignable attributes, or have it all hard-coded on level up?

What time frame should we shoot for?

Should we have vehicles...?

[pixel_legolas]

well if we are going for a good old FF style we set the stats from beginning, let the character evolve with levels and have +stats with different weapons.

As I have not done any projects before it's like jumping into a dark and deep ocean so I want to start small. Maybe 2 characters to select from first with different stats. A couple of enemies to fight against and a boss battle maybe?

If we could just get a basic system working we can then add everything, but the basic system will be the hardest part also

[zknack]

Things needed to be done:
*Create a basic moving area map
*Create 3-4 Characters, even if they're lego men for now (2 main character options, I'd say a male & a female, and 1-2 party members)
*Decide on what stats we want to have
*Decide on how to work the combat mechanics, as there's multiple methods.
*Create a basic, clickable GUI for combat *only*
*Modify Save Game file to hold the stats, info & character name
*Create a stand-in Clickable menu to see 'party' and 'items'
*Create 'party' and 'items' menu that reflects at least a list of items found/held
*design 1-2 combat maps
*ensure to lock movement in combat field when teleporting there
*Create hand-holding file for the terrain to have the terrain know what type of combat map to send you to and
type of enemies there.
*Create 1-3 enemy types (may have this covered, but they're not rigged well)

Things done by the engine or me already:
*Teleporters for teleporting player to combat map
*Movement (yay!)
*Examples for damage (Kinda yay!, although not quite what we are looking for)
*Plays animations
*I have a ******ton****** of mocap data that is legitimate & owned, many that would work for this.
*I have a few swords and staves already waiting for such a game.

Am I missing anything?

[pixel_legolas]

I have a friend who made alot of "one piece" models, the anime if you don't know about it. I can ask him if I can modify them to new characters with new textures. He has already made some combat animations for some of them.

We need some sounds too. Some smooth background music and some footstep sounds. Maybe a vehicle in the world map.

There is one thing that i just thought about, totally random. The reason why I don't have made anything solid is that as soon as I get stuck after tying some things I post a question on the forum, it takes such a long time for answer so the energy I had just leaves. I wish we could have dedicated programmers for GC who is more active in forum, working for helios/gekido.

This would keep up the energy-flow for a project.

Maybe we can start showing pics of some art assets. We need grass/stone textures, trees, stones and maybe some bushes. Perhaps a small pond, let's start easy.

Of course we need a chest that can open and spit out a herb or sword

[zknack]

Grass / Stone Textures I have a few of, as well as a couple of trees rocks, & bushes.

Music......... Hmm... I have a few from buying the arteria3d sub.

Some of the assets or asset libraries I have that may help for this:
* Entire TrueBones library (holy bejeebus a lot of animations)
* Arteria 1 year sub for sound/animation/models/etc (dang good deal!)
* Did the Beyond Virtual / Game Core class & have an active subscription for updates w/them
                  \-----------------> Can't share the framework, but can PM details of what it contains that can help that
                                            I could use to plug in.
* Tree Creation Tools

------------------------------
If your friend is willing to share some models that would be awesome. I have some low poly models from a project I was working on for 2 years approx., that are very, very fantasy-type (bridges, rock formations, some houses, 2 windmills, a lighthouse or two, a few ships, a wagon or three, common items, skeleton warriors, orc type creature, treasure chests (some with animations), a good bit of weapons- most made with a sword creation toolset that was going to be for rgen media/ rpg builder- but they more or less are gone/too flakey to
work with.).

[Squat]

Well, I wish I could be around more often to provide support to others...hell I'd even spend a day trying to code through their issues myself. But the truth is that life is a drain and I can't do that while I gotta worry about finding a job and all that stuff. It doesn't appear they're interested in aiding the developers in any sort of meaningful way, and who can blame them. They got their own mouths to feed. Until the world is perfect and we all help each other, I hate to say it but you're gonna be about 95% on your own on just about everything you try to do...just like I am.

[zknack]

Squat,

I hope things get better for you man, I miss your more ingenious, upbeat self.

If there was anything coding-wise I could help you on I would, but you're light
years ahead of me there.

I'm still (as before) willing to aid you on models/textures if you give me a good
idea on what's needed.

[pixel_legolas]

I think we should open up a community project with open directory in some way so we gather files in a ftp, or something, we just paste new stuff and make the game evolve. The hard part would be to make sure people are not writing over things that are already newer.

I experimented with modal dialog yesterday and seems quite easy to implement. Could script a "tutorial for beginners in X game" so people learn the basics of the game etc.

[gekido]

Was thinking about this exact thing last night! 

Haven't done much more than small experiments for specific turn-based implementations, but the basic structure of a turn-based game wouldn't be too hard to implement.  I have done some implementation of character-creation / rpg-style stats & 'party' creation though.

One big thing that I will point out with regards to ANY kind of rpg / stats-based game:
- The save game system is NOT the only mechanism you have for storing / saving information.

All the SaveGame system does is write an OptionsFile - but it is very restrictive, ie you only have the one save game file to work with.

Using the OptionsFile system, you can have hundreds / thousands of different config / stats files that can be read as necessary depending on what is happening in your game.  The best way to think of OptionsFiles is that it is a very robust config file parser that can be used for whatever you want.

To use an example:

the ContentBrowser mini-app that i'm working on has dozens of weapons, each with their own unique OptionsFile.

The weapons system as a whole has a high-level Master List that tells the browser which weapons are active & should be available in the game.  This master list is ALSO an optionsfile.

Same thing for vehicles, units, characters, squads, ammunition types, weapon effects, etc.  Literally any kind of 'table' of information (thinking of a traditional RPG), can and probably should be stored & read via an OptionsFile.

================

Now, specifically regarding the 'turn-based' nature of the gameplay, basically you just need a flag for each individual unit / object (a ControllerVariable) that tells the unit that it's 'their turn' to move.  If it's not their turn to move, all they should do is process 'reactionary' type events - for example, getting shot, etc.

From the player-perspective, you'd probably want to do something similar to an RTS game - which I 'have' been experimenting a bit with.  Essentially how the structure works is (for an RTS game anyways) something like this:

Camera
--- AI Controller
------- Character Object

The camera is the high-level controller that handles user input, gives directions etc.

The AI Controller provides the basic 'brain' for the character - ie if there is an enemy nearby, attack etc.  This way you can reduce the amount of micromanagement that the player needs to do.

The character object is the actual visual representation of the unit in-game.  it receives 'MoveToPosition' commands 'AttackEnemy' commands and so on from the player (via the Camera object) and does the low-level movement / animation control.

This way you keep the various logic layers seperate - makes things much easier to understand and process when you are working on the scripts.

=============

All you'd need for a turn-based RPG system is basically a layer that has a mechanism to:

1) send a message to a unit to indicate that it's their turn to move
2) query the unit to see if they have used up all of their movement points for that turn yet

I have a functioning version of the above RTS system that I can post up (using the lego dude). It does the basics of a top-down RTS style game, including unit grouping, selection, movement actions etc.

Obviously there's alot of work that would still need to be done, but it would be a starting point at least?

[zknack]

Gekido,

That would be superb! The RTS chunk would be amazingly helpful, as would any more info you care to
share about how to integrate the optionsfile.

I think a basic free framework for an RPG and RTS would create a good amount of interest here, and
undoubtably generate more interest in the system.

My thoughts prior to your post was to simply enable / disable the GUI for combat untill the last enemy
had acted/or reacted thus far, but with some sort of example as you've suggested now, I can see it
becoming more flexible for further use.

I would not mind in the least making a chunk of content that I have made & use for this (if it becomes
a community thing) freely available.