Friday, August 03, 2012

Gripping Turn Based Combat

Turn based combat bothers me a little. See, it originally evolved for war games. Giant maps in your basement covered in little tanks and fake trees. The whole point was that it allowed you to play a game with waaaay too many moving parts and rules.

Now we see it in many other places. For example, computer RPGs. So we're using a system developed to keep details under control in the absence of a computer... on a computer that can keep details under control anyway. That's the definition of irony, according to a song I once heard.

Today I was also thinking about gamefeel. Turn based systems are usually about choosing which tactic to use. Considering how awesome it feels to take a turn is normally not high on the list of considerations. But what if we were to do just that?

What could we come up with if we wanted to do a sort-of-turn-based combat system while focusing on it being responsive and fun to use?

It's an interesting challenge. The play has to respond to the player's touch sharply and tightly, but you have to have depth and skill involved. The player has to be able to see exactly what is going on, and his actions should become second-nature enough that he does them automatically to express his wishes, rather than having to think about traversing menus.

I'm also arbitrarily writing off the idea of a detailed grid map, mostly because I think the grid map is too tightly linked to our concept of a turn-based game, so using it would limit our creativity.

I think an answer is in the initiative system. I think everything can boil down to initiative. Who gets turns, when. Moreover, how many turn-order phases your action can execute for before being interrupted.

Imagine that every battle's centerpiece was a giant turn order wheel. You could clearly see who gets to go, when. Let's say Anna is fighting two goblins.

Anna sees herself, and she sees the goblins going after four turn-order phases on the wheel. She has a few options, such as "attack". She mouses over attack, and sees an arrow from now forwards in time on the turn wheel, shooting past the goblins and a few ticks further. Simple, right? On the goblin's turn, they do much the same. Press attack, choose a target, watch it resolve, watch the wheel spin.

Well, on her next turn she decides to do "flurry", a more advanced attack. The arrow once again flies forward on the turn wheel, but this arrow is not properly an arrow because it does not end. Instead, every two or three turn-order phases there is a hack mark and it fades out after twenty or so phases to indicate it will continue until the chain is broken. Once she selects this and selects a target, she will automatically attack that target every time there is a hack mark until she is interrupted or her target dies.

Barry is a knight, and he comes in specifically to help her with that. Barry's ability is to intercept attackers. He selects "guard" and then selects her. Now, when an enemy attacks her, Barry is in the way. Moreover, Barry's next turn happens the turn phase after he intercepts an attack, or the turn phase after Anna's ongoing attack stops.

Alternately, Chad might cast his "freeze" spell, which has a long refresh rate but causes a target enemy's turn to be delayed for quite a long time as well. There are plenty of other tactical options - tradeoffs as you manipulate the wheel.

I think this has some merit. It's very clear what is happening, when. There are a lot of fun tactical tradeoffs, and the control is easy and, if properly implemented, crisp and clear.

... Maybe I'll even do a demo of it, if I can drag myself away from the demos I'm already doing.


Adam Augusta said...

Reminds me just a little of Puzzle Quest; spending time and effort trying to generate or suppress initiative, with actual actions changing the environment in a way that could significantly modify the decision tree.

In the case of Puzzle Quest, though, that dynamic only worked for a very limited stretch of the game.

Craig Perko said...

I played Puzzle Quest. All I remember is match-3 stuff.

Adam Augusta said...

Match-3 was the basis for generating/suppressing initiative ("mana") for board-changing actions. It might have been one of the sequels that did a better job in this regard, but only, again, for a brief time before the game just got kind of degenerate.

That said, I don't want to be that guy who's all like, "Your idea's like this thing!" when it's not like that thing at all. I just want to chime in and say that making everything about initiative seems really, really fun. And rare.

Craig Perko said...

When it comes to game design, it can be valuable to look at an idea as it was implemented in other games. However, the downside is that if you describe something and someone has played a game with a similar mechanic, they can only see the mechanic they played, not the one you are actually describing.

Craig Perko said...

(Not saying that's what's happening here, just that it tends to be what happens.)

Random_Phobosis said...

To jazz it up further, maybe I'd allow the characters to take multiple actions each turn if they like, but each "action slot" could require more mana for ability use and/or push the character further on the turn order and/or impose penalties for defense. So a character could finish off the enemy if absolutely necessary, but at the cost of vulnerability. That's usually what the grids/map locations are for anyway.

By the way, totally offtopic, but for me both Puzzle Quests were games about engine building, where you try to combine abilities into some kind of unstoppable combo chains before the battle, and then just watch your plan unfolding.

Craig Perko said...

Hmmm. Maybe there are "chaos" type abilities which can be used on your turn without spending your turn, but instead push back all your allies' turns.

Adam Augusta said...

I assure you, that *totally* happened here.

You presented an experiential riddle. I was unable to unravel it, but I found an analog, and my "fast brain" is convinced that it maps well. My "slow brain" is convinced that it doesn't, but is willing to allow for the possibility that the analog might be interesting. My "fast brain" decides that the risk of embarrassment or censure is outweighed by the joy of being a participant, and so the comment is made.

I love people.


Antsan said...

I've only read the first two paragraphs, so...

I think you might want to read
The writer has a very precise idea of a what a game is and isn't and he has a very good point about why games normally *should* be turn based. According to him games are about choices, not about applying your them. Real Time is always about timing or cinematics or simply unquestioned convention and thus not really helpful for making games.

Craig Perko said...

I agree that you can make turn-based systems interesting. Hell, the result of the essay is a turn-based system.

That said, real-time games are a lot of fun, too.