Developing an RPG - opinions needed

Started by
26 comments, last by wirya 16 years, 1 month ago
Quote:Original post by wirya
3. More than 100 characters, each with it’s own unique background and characteristic.
5. More than 30 types of creature, each with it’s own characteristic.

I might describe the features in details, if you're interested.


I'm curious what those characteristics are. I hope it's not weight or color, but things like personality and behavior? And if yes, I'd like to see what you have planned for those 100 characters and 30 monster types :)
Advertisement
the idea's for a game sound interesting enough

but, they wont really make a game, only giving me idea's of types of games you could make with those features

Sell the idea more, how will it look like a normal application?
like drug wars, or more eye candy then that?

Pricing is a bit of a meh issue, if it looked applicationish
and not online? (which seems a bit strange to have so many classes/roles)
id make it free, put a add banner along the bottom, sell as adware.

What id like to see, is NPC's that do progress, there not forever waiting around for you, like that oblivion game (though i couldn't get into it, wasn't enough change that i could see that directly affected me)
Which does seem to one of your goals.

You need more information, then a list, maybe take 10% of each number and describe indepth about them, what would make npc 1 diffrent to npc 5

Im sure you get the idea ^.^
Don't mistake Complicated for Complexity

A complicated game has lots of options but relationships between those options are not deep. In a complex game the relationships between the options are deep.

For example, in chess, you have 6 different pieces (Pawn, Rook, Knight, Bishop, Queen and King) and can place them on one square on a board that is 8 x 8 squares. At any one time you might have a maximum couple of dozen choices (usually it is less than half a dozen vaild/worthwhile moves in practice) as to where and what you can move. So compared to the list you provided for your idea, Chess is not very complicated.

However...

In chess, because of the relationships between the pieces (how they can move and where they are possitionsed), it becomes an extremely deep game. In fact, modern computeres can't even calculate every single interaction of the pieces. It is that deep.

Another thing aobut complexity and complication: Humans can only handle so many choices before they can't handle any more. Humans seem to max out at around 7 options, with a typical number being around 5.

So having a choice of around 25 classes could overwhealm most players.

Howerver ther is a rescue. It is called "Chunking".

Chunking is where someone mentally groups objects or choices together. This enables them to compare many more choices at the same time.

For instance:
If you split your classes into 5 main types, each having 5 varients in each type and make this clear to the players, then this will allow them to Chunk the choices into the limit of 5.

First they would decide between the main types. Maybe these are: Striker, Defender, Ranged, General Support and Fire Support (note the might not be what you call them for the player but are just to help you understand how you might chunk them).

Then they would choose one of the subtypes, for example under General Support you might have: Healer, Buffer, Debuffer, Aggro (gets enemies attention) and Stealth (open locks, disarm traps, etc).

Of course in game you might list these as Cleric (Healer), Enchanter (Buffer), Abjurer (Debuffer), Aggro (Invoker), Rogue (Stealth). Or whatever fits you game world.

By breaking it up, you make the choice easier for the player, thye might decide that they want to play a support role character, then they can choose one of the support classes (say an Invoker).

As you might be able to see, by chunking it this way you automatically increase the depth of the game. You make it less complicated, but make it more complex.

There is more than jsut reaning this though, when you begin to chunk this way you actually limit your choices as a designer as to what you can do. It kind of restrict your thinking, but it can make a better system in the end. For instance, under the chunking it would be hard to have (without allowing multiclassing) to have a Support that was also a Striker.

Even if you don't buld chunking into you game, player will automatically do it if they get enough experience (in fact with chess, you can roughly determine how good a player is by how much and how well they chunk the game).
Thanks for the replies, guys. I'll try to give the best reply.


About this project of mine. The goal is to develop a TRUE RPG. So, the player will really role-play. And since I believe that what defines a role is the game action set it has, so my design is focused on making a strict and efficient game action system (and therefore I'll also have to make a strict and efficient information system and other things).


Quote:Original post by Kest
An application? That's about as minimal as it gets. I would be drawn to any well designed and balanced game, as long as there is some type of representation beyound buttons and text. It's difficult for me to even imagine real time action with buttons and text.

It'll have 2D pictures of the game objects (characters, creatures, weapons, rooms, etc). Hand-drawn. There'll be no animations. That's what I meant by "minimal eye-candy".


Quote:Original post by Kest
To be completely honest, you didn't describe much at all about your project. Most of the features you listed are a part of any role playing game. The sheer number of types of objects and actions is not something that's very important to real gamers. What is the point of the gameplay? To survive? To grow? To conquer? To save? What will the player be doing the most of to accomplish that goal? Fighting? Conversing? Planning strategy?

Well I posted the features just to know what the reactions of people would be when I advertise it :)


Quote:Original post by Kest
Isn't it a little early to decide that? I mean you'll be working on this game for about 4-8 years. The stock market may collapse again before you finish.

Well, I've developed some parts of the game, and my estimation the game will be finished before 2009. I have an early version of it with the early interface (looks really ugly), it can't be played yet (buggy as hell) but it shows what kind of game I'm talking about. If you guys are interested, I'll upload it.


Quote:Original post by RivieraKid
What is it exactly which sets your game apart?

At least these aspects :
1. real-time action system, the actions will have specific STARTUP time, ACTIVE time, and RECOVERY time. Action timing strategy will play a great part here.
2. EVERYTHING you and the AI do will be a full game action, meaning that whatever kind of action it is it would be treated the same. Even the talk and trade actions will have STARTUP, ACTIVE, and RECOVERY time. So for example, there won't be a Morrowind-like talk action which when triggered it'll put the game world to a halt (you can't get hit when the dialog box is open).
3. Classes/roles which forms a solid community/society. You're a warrior and want to get your wounds treated? Go ask the healer. You're a healer and want to get some money? Tell people that you're a healer, wait in your shop (well or you can go wandering around), heal people. You're a knight who command an army of warriors and want to strike the badguys? Call the warriors, tell the healers/mages/others in the city that your army need their services, and strike 'em. You're a humble trader who happens to know that the badguys are going to kill the leader of the nation? Go and tell the royal army commander about it, and then do whatever you can to help 'em (gathering weapons, potions, armors, joining the army, etc). And such.
4. This hasn't been mentioned. There is no leveling system. The purpose of the game really isn't hunting for EXP.
5. Real-time information distribution system. You spread informations using TALK action, you gather information using LISTEN action, you can even read people's mind using spell, or use TELEPATHY action to avoid people using LISTEN action on you (of course not all classes have this action). All those actions are, like I said, full game actions, they have STARTUP, ACTIVE, and RECOVERY time, so for example you can get interrupted before you finish your TALK action. And the informations themselves would be various, it can be about alliance, about who hate who, about where a specific person is, and even about how much that particular item worths, etc.
6. The PC (Player Character) is only one at a time (you choose it from the beginning of the game), but you basically can play as any of the 100+ characters (including the badguys and even the leader of the nation). Each of the characters has a fixed class, meaning that if you want to play as another class then you'll have to switch character. Of course that (one PC at a time) doesn't mean that you'll be somewhat lonely, because you can always ask for help from the NPCs (especially if your class/role is "Nation Leader" :D ).


Quote:Original post by sunandshadow
25 roles/classes seems like an awful lot to develop each in any depth. I'd rather see a game with 4 unique deeply developed classes or races than one with 25 that you can barely tell apart.

Well, it's actually not that hard to develop the classes. The actions in this game can have up to 4 effects. So for example, a mage can have an action which have PHYSICAL ATTACK effect, just like a warrior. But a warrior will have more various attack actions. Warrior can have a quick attack action (low damage but also doesn't require much stamina), a heavy-damaging one, a fake one (has a really quick STARTUP and RECOVERY but zero ACTIVE time), and such. Warrior can also have an action which have not only PHYSICAL ATTACK effect but also PHYSICAL GUARD effect (in this game it'll be called "Warrior's Charge"). Mages won't have that kind of action.

And since there will be more than 70 action effects (some of them are : PHYSICAL ATTACK, ENERGY ATTACK, CAST SPELL, TALK, LISTEN, MEDITATE, REST, READ, GO THROUGH, EXAMINE ROOM, EXAMINE WEAPONS, EXAMINE MENTAL, EXAMINE PHYSIC, HIDE, TRADE, CREATE ALCHEMY), developing 25 classes really isn't that hard :)


Quote:Original post by Hajo The Dreamer

I'm curious what those characteristics are. I hope it's not weight or color, but things like personality and behavior? And if yes, I'd like to see what you have planned for those 100 characters and 30 monster types :)

For the characters, it'll be : clan (allied clan, opposing clan, rival clan, high clan, lower clan, etc), attitude (cowardly, brave, distrust foreigner, greedy, etc), goal in life (opposing government, helping government, neutral/go with the flow, etc), and such. And of course the roles/classes would have significant effect.

For the creatures, it'll be : attitude (tame or not, hard to scare, easy to scare, hate specific creatures, hate specific person, etc), intelligence (stupid, cunning, etc), and such.

No masher just Master!
Well... I have to applaud the goal of an RPG where people actually roleplay, but I have to say I don't like this design or think it will accomplish your goal. A rigid class system is anathema to roleplay no matter how many class choices you give the player. A much better idea would be to have no classes, but instead award the player class labels based on how they build their character (either by which abilities they choose/purchase from the master list to build their character, or even better by what abilities they use most in the game (crafting, stealth, shapeshifting, damage blacking, raw firepower, long range, stunning/paralyzing opponents, maybe even building trap...), and give them bonuses which suit their build, quest rewards or unlocking new abilities (as well as customizing appearance lol). Make it free and easy for the player to rebuild their character at any time, or make a max level character able to afford/have all abilities.

Then if you have time and energy left you could put your categorizing and detail into race/species choices which would each have unique personalities and quest lines.

I want to help design a "sandpark" MMO. Optional interactive story with quests and deeply characterized NPCs, plus sandbox elements like player-craftable housing and lots of other crafting. If you are starting a design of this type, please PM me. I also love pet-breeding games.

Quote:Original post by wirya
About this project of mine. The goal is to develop a TRUE RPG. So, the player will really role-play. And since I believe that what defines a role is the game action set it has, so my design is focused on making a strict and efficient game action system (and therefore I'll also have to make a strict and efficient information system and other things).
I don't think you can do that in a PC-RPG. A human brain is required to process this kind of complexity.
Quote:Original post by wirya
It'll have 2D pictures of the game objects (characters, creatures, weapons, rooms, etc). Hand-drawn. There'll be no animations. That's what I meant by "minimal eye-candy".
I don't know why but I find this interesting. Like an evolved MUD. Sounds nice.
Quote:Original post by wirya
3. Classes/roles which forms a solid community/society. You're a warrior and want to get your wounds treated? Go ask the healer. You're a healer ... You're a knight ... You're a humble trader ...
Uhm, this sounds like lot of work. Implementing this isn't going to be easy, the mechanics are way complex and unless you script everything (implying there would be only pre-defined scenarios) this feature alone would be overkill.
Quote:Original post by wirya
For the characters, it'll be : clan (allied clan, opposing clan, rival clan, high clan, lower clan, etc), attitude (cowardly, brave, distrust foreigner, greedy, etc), goal in life (opposing government, helping government, neutral/go with the flow, etc), and such. And of course the roles/classes would have significant effect.
In low level terms, the "clan" or "opposing clan" property isn't significantly different from the "color" or "weight" property. I suppose each clan is going to be a runtime token and a few checks are going to be run on sight or higher-level routines.
Take for example the "goal in life" property. This won't come out of silicon by itself and this implies there will be some kind of switch somewhere... have fun debugging weird behaviours that shouldn't happen (Neverwinter Nights' "disappearing orcs" bug is historical)!

I understand this is a "design" forum but when you write your ideas, do you take in consideration the work required to implement them in a way emergent complexity is considered "interesting"? It's part of the design as well. Putting togheter all your ideas isn't trivial.

Previously "Krohm"

Quote:Original post by sunandshadow
A much better idea would be to have no classes, but instead award the player class labels based on how they build their character (either by which abilities they choose/purchase from the master list to build their character, or even better by what abilities they use most in the game (crafting, stealth, shapeshifting, damage blacking, raw firepower, long range, stunning/paralyzing opponents, maybe even building trap...), and give them bonuses which suit their build, quest rewards or unlocking new abilities (as well as customizing appearance lol).

I agree. Earning a title is also much more rewarding and meaningful than choosing one from a list when the game starts. You could still give the player class-like rewards or bonuses after they reach certain titles. To make it even more interesting as a gameplay concept, the number of classes (or titles) should be hidden from the player, as well as the details of how to obtain them. You could have very unique classes that require extreme specialization to obtain, giving players a lot of incentive to experiment, and raising the replay value considerably.
Wirya , have you ever programmed any video game before ?
I'm currently developing a 2d game with animated characters.
I have a powerful machine with latest video card and tech.
And when I try to select many entities ( >50 ) the frame rate slows down.
My project uses only Direct X/C++ , no SDL only mere Direct X...
It means that I must afford a lot of code optimization...

I would suggest you to start a simple game before considering a sophisticated one.
Can you give us the proof of your successfully finished game ?
Quote:Original post by HolyGrail
I have a powerful machine with latest video card and tech.
And when I try to select many entities ( >50 ) the frame rate slows down.
My project uses only Direct X/C++ , no SDL only mere Direct X...
It means that I must afford a lot of code optimization...

Code optimization will only help a very little. Most of the dramatic improvements of video game performance is in finding clever ways of reducing the workload. Not code-wise, but logic-wise. For example, a grid that allows objects to only perform expensive AI tests on enemies that are very close will yield huge improvements when there are many enemies, where optimizing the AI tests themselves will barely be noticable.
It like it, except for all the numbers. The real time compound action system things sounds cool I've though about doing something similar.

I dont see anything wrong with classes, they help players to easily decide what they want to do, and its just a simplified reflection of real world limits. It all depends on how you implement it.

This topic is closed to new replies.

Advertisement