Hi. Hopefully, this will be the first of many posts as I work my way through my first major game project. I've done Asteroids and Bomberman clones before, but I think I've honed my skills enough now to take things to a higher level. The project I'm working on now is a two-dimensional RPG in a cyberpunk style setting. I've spent a good bit of time writing the script and am now in the process of deciding on the gameplay mechanics. Each character has a list of skills, some of which are combat oriented (ex. one character is a highly accurate gunfighter, another is a kind of street boxer), while others are used outside of battle to solve various problems (ex. computer hacking, lock picking, etc). Overall, there will be more of a focus on problem solving, puzzles, and the like, but I still need a solid battle system in place for the times when a fight is necessary. The old fallback would be the traditional turn-based fight, and in some ways, that might be the best option, as it's easy for the player to understand and allows easy access to various skills. However, such a system runs the risk of killing the fluidity of the game with the transitions into battle and back. Chrono Trigger did a decent job of making a fluid transition, but there was still a bit of a start-stop nature to the game. The other option is to go for more of a Secret Of Mana style action game approach. While battle and non-battle scenes did mesh well together, allowing there to be instances where the puzzle elements intermingle with battle elements, the majority of these games are rather clunky when it comes to ranged weaponry. In many of these games, ranged weapons are either pitifully weak or short ranged, and in many instances, you end up getting shot by enemies that are far offscreen. In a game that primarily features ranged weapons, with only a few melee fighters, is it possible to balance out such a system without placing unreasonable restrictions on the player or AI? Any advice would be appreciated, subjective or objective, as would any well-written articles on the subject. Thanks a lot.

Since this is a personal project i would say stay away from turn-based simply because when doing your testing/development, it just gets sooooo boring waiting for your turn. Even for players, i think it is really boring and battles are not fun and become a drag. I remember i used to hate battles in Final Fantasy when walking around hoping to find out what happens next in the story.

Ironically i used a turn based system for my FYP bcos i felt that it was the easiest to do compared to Action-RPG Ala Diablo. I could do with simple to no AI which can be quite difficult for a beginner. Turn-based systems are easy to implement but extremely tedious because of all the menus/icons and rules. On the other hand, action-RPGs are much harder to get right but require less amount of work.

Quote:

Original post by GamerSg Since this is a personal project i would say stay away from turn-based simply because when doing your testing/development, it just gets sooooo boring waiting for your turn. Even for players, i think it is really boring and battles are not fun and become a drag. [...] [...]Turn-based systems are easy to implement but extremely tedious because of all the menus/icons and rules. On the other hand, action-RPGs are much harder to get right but require less amount of work.

However, the two (real-time vs. turn-based) aren't really that simply comparable because they are fundamentally different on many levels.

Turn-based combat is based on the concept of tactics and player's intelligence. There is really no limit to the detail you can put into the combat mechanics, as the player has infinite time to make his decision on what to do (well, you could put time-limits on the turns, but that's not the point right now).

Real-time combat, on the other hand, is based on action and player's reflexes. You have to make a lot of simplifications since if the players had to decide between a hundred options every second, many would probably just quit the game. Remember that real-time combat can induce a very strict relation between the player's skills and success in the game (since timing can become crucial), so if you want to emphasize the separation between the character and the player (in this case it means that it is the character who fights and the player's reflexes shouldn't have such a large effect), real-time combat will make that difficult.

I'd even go as far as to say that a turn-based would inherently imply a much more abstract feel to the game than real-time; at least if you put in a lot of simplifications. People will probably forget more easily the fact that a sword is not just a damage value distribution in a real-time game than in a turn-based game as they would have less time to think about it... Also, real life doesn't happen in turns. You don't wait for the enemy to hit you first before you hit him. Then again, you could make it so that instead of taking turns in an alternating pattern, each player (incl. computer controlled characters, monsters, whatever) first tells the UI what they want to be done on the next turn and then all the entities in the wolrd move simultaneously. It'd still be turn-based, but not with the silly "ok, now it's your turn to hit" mentality.

Of course you can make all sort of hybrid implementations. By having the possibility to pause and give orders in the heat of battle, you basically give the possibility to use tactics but at the cost of killing the flow of the game. Then again you could have a time-limited turn-based system that was displayed in real time, as in e.g. Neverwinter Nights. Fallout Tactics had each action cost action points and basically the flow of the game was divided in microturns of sort during which an amount of time equivalent of one action point passed. Or something like that.

Turn-based can be tedious (then again, no one is telling you to put in hundreds of menus and rules in turn-based systems either), but quite similarily, real-time can be too hectic. Drinking half your rejuvenation potions when you panic and just slam the keyboard at the sight of having only 1 hit point is no more fun than falling asleep waiting for your turn.

Who are you making this game for? What is the purpose of combat? If you want to make it tactical and test the player's intelligence, make it turn-based. If you want to make an action game and get the player's adrenaline flowing, make it real-time. Of course this is just a general suggestion and I'm not saying turn-based can't get your adrenaline flowing or that real-time couldn't be tactical.

And if you want to make it multiplayer you have to consider things like that the player will have to wait for the other players to take their turns and that can take a long time, depending how many players are playing etc. Still, it can be fun if the turns don't take too much time (as we are not talking about a strategy game with hundreds of units here).

As with many RPGs, one of the main foci in this game will be the story, but overall, I want to encourage the player to use his head. At times, there will only be one way to progress through the game by using a specific character's skills (like using the strong character to move a heavy crate out of the way). Other times, there will be multiple ways to get through an area, and your choice will have different effects (if you break down a door, a guard walking by may sound the alarm, whereas if you pick the lock or sneak through the air duct, they will not notice). Some decisions may yield rewards such as money or items. In some cases, it may be possible to avoid combat by making the right decisions, whereas some times, it will be unavoidable, whether it be because you get attacked by thugs while walking down the street, because you need to stop the big bad guy from carrying out his plans, or because stealth simply doesn't apply to the situation at hand. It may even be possible to complete the task by smashing everyone and everything in sight, but it will obviously be more difficult if that isn't the optimal decision. I hope this will lead to some interesting decisions on the player's part in the times when he can choose his party, as some characters will be more combat oriented, whereas others have skills to help avoid it.

EDIT: I'm sorry... I just realized that I didn't really do much to answer your question there. Given the suggestions you've provided so far, I'm leaning toward a semi-realtime system where you control one character directly, and the rest of the party is mostly controlled via AI, with the ability for the player to issue commands to them as necessary, much like in SoM, Seiken Densetsu 3, or to a lesser extent, the Tales games. This would allow seamless transitions into and out of battle and maintain a semi-cinematic feel. Once I get a basic framework down, I can move on to the math... Would anyone happen to know of any good articles on stat-based calculations in battle? I mean, the actual math would be simple, I'd imagine, but using the right calculations and assigning the correct numbers to each character and enemy stat will have a huge impact on the balance and challenge of the game, and I'd imagine that simple trial and error would be a horribly inefficient way to attain balance.

[Edited by - Davian on March 22, 2005 2:38:14 PM]