Tactics fans: What do you think of turn order?
I've mentioned a few times I'm working on making a tactical RPG engine, and I'm finally at the point where I'm doing the actual important part of determining turn order.
It's a harder design decision than I thought, so I thought I'd ask for opinions.
I'm trying to make a game in the style of Shining Force, where the turn order is determined by a unit's status. There's quite a few games like that, but I'm wondering if that design is actually good. The thing that makes that style of game good is that they're actually fairly easy, and the "noise" of stat-based turn order, where you can't depend on a specific turn order, seems to make the game harder. Or at the very least, it closes the door on more precise tactics. I've had plenty of experiences where I thought a unit was safe because they had enough HP for one more hit, and the enemy ended up hitting them twice before their turn came up.
In reality, I'm actually pretty set on keeping this in my engine for better or for worse (boy, is it hard to resist expanding scope), but I'd be interested in hearing some opinions nonetheless
I'm going to talk about the turn order mechanics in Divinity: Original Sin 2 (absolutely not a tactics game, definitely a more traditional RPG). Not just because I love the game (and will take nearly any opportunity to talk about it), but because I find its initiative system interesting. Note I said interesting and not good. It has issues, but I like it enough as another strategic element within the game's systems.
Initiative is stat based. Each point in Wits gives you one point in initiative, and random gear can have some amount of bonus initiative based on rarity. Higher stats means go earlier. Except that's not what happens because of the first interesting thing that D:OS2 does.
Initiative alternates between teams. The character with the highest initiative overall will still go first in the round (barring scripted first round initiative on some fights), but the rest of the characters will alternate. Say you're in a fight between your party (A) of four people and an enemy team (B) of six. Let's say that one of the members of your team has the highest initiative, which then means the initiative order will look like ABABABABBB. If one of the enemies had the highest overall, then it would be BABABABABB
This system has a couple of consequences:
The most interesting thing to me is a little button to the right of the "End turn" button: Delay. If you haven't taken an action in your turn, you can hit the Delay button to put yourself at the end of the initiative order. Say the second person in your party hits delay; the turn order for that turn will look like ABBABABBBA, as they've moved from their spot to the end. What's more interesting is what happens when multiple people hit Delay, as you quickly find out that the delayed turns happen in reverse initiative order. I suspect the reason for this was pragmatic: it prevents players from being in an endless loop of delaying past each other.
Delaying becomes a real strategic decision: you're giving the enemies an opportunity to attack and possibly freeze/stun/knockdown/charm you, but in return you can co-ordinate moves more easily with your party members. My favourite thing to do was ensure my character had the highest initiative in the fight (the extra crit chance from Wits didn't hurt) and delay my turn to the end of turn 1. I'd then have my turn at the end of turn 1, but since my character had the highest initiative they'd be right at the top of round 2, allowing me to set off some combinations of skills that would have been riskier if my turns were a full turn apart.
The reason I bring this system up isn't because it's the best thing in the world. Every time I've played the game with a friend I've had to explain the quirks of the initiative system and why the enemies are getting so many turns in a row (it's because we're outnumbered and they all have to go at the end. You could have delayed your turn but that would have put you at the end and you wouldn't have teleported Gwydian Rince out of the Necrofire and back into the Dome. For the second time. I forgot to save again, so if we reloaded we'd be back at the start. Yes, those ones are immune to fire. Uh, sorry. Just had a flashback). I bring it up because two small decisions (alternation and delaying) brought a new dimension to what is a pretty staple feature of RPGs. There are ways of adding mechanical depth to the system without ripping it out or overhauling, but since it is so integral it's hard to get right.
I like being able to manipulate turn order but maybe to prevent your scenario everyone still gets to go at least once before someone can go a second time. So using haste or light weight weapons or however you want to think of it just makes you go faster in that "round" but all of the slow people still get their turn.
Then I would say the only way to get more than one action in a turn or lose an action in a turn is through doing something active. Casting a spell to go twice. Attacking someone to interrupt a spell. I think that gives a good layer of strategy. Do you try to finish someone or do you try to interrupt someone who potentially has a big attack coming.
I like this status buff approach too, though it's possible to rig the experience if you have items or spells that extenuate these traits too far.
Like, wasn't Brave and Faith in Final Fantasy Tactics an example of breaking the game and the turn order? I'm having a hard time remembering.
I'm playing the first episode of Shining Force 3 right now and purely from a playability perspective, I really like the balance created by weaker players (defense) appearing earlier in the turn order and having to err on the side of caution with my moves for them as enemies might try to take them out before my tanks arrive.
Very cool @Akir, that you're making a Shining Force style S-RPG. Playing SF2 as a kid was an amazing, age appropriate introduction to strategy and storytelling.
I know about three main types of turn orders:
Speed or RNG order, but round robin
Speed or RNG order, not round robin
Fire emblem.
isn't all that common in video games, but it is a staple of board games and TTRPGs. After all, it would suck if someone else got to play 2x as much. I think it works the best if the units are very powerful and unique - like a DnD party. Otherwise, the fact that everyone has to at least go once limits the strategic depth.
is pretty common in video games. I think the main thing is that it can drag a bit. Additionally, sometimes it's hard to actually string together unit moves.
is mainly fire emblem. I honestly do enjoy it, because it's so fast paced - you can even just skip showing the enemy turn. But strategically, it's kinda fucked up. Since you can move everyone on your turn, you can do an immense amount on that turn. It means that you're incentivized to play defensively - whoever is attacked first, will get mauled. It also allows you to do pretty cheesy things, which you can see on any fire emblem speed run.
I'm guessing that numbering was meant to be 1 2 3, 1 2 3 before formatting ate it, right?
The other weird thing about Fire Emblem is the counterattacks, because certain games - depending on the AI, the disparity between party units and the enemy, or stat calculations - can make the enemy phase more advantageous for you. In a game where the enemy stat disparity is obscenely high, like 8, your dodge-tank Swordmaster can only attack once on their turn but can hypothetically dispatch an infinite amount of units walking in a single-file line to attack them on the enemy turn. When the enemies are more dangerous, it's way slower and more tactical. I'm playing through Shadows of Valentia, and even on Normal difficulty I'm finding though I don't get in a lot of tough binds I still believe it's more thinky than a lot of the other entries. You can't just leave most of your units exposed unless you put all your buffs on one unit; a bunch get doubled frequently and some will crumble to 2-3 hits from the wrong defensive type.
Point for OP being - whatever you do, it's inevitably going to make a big change. If you do a stat-based initiative, I think you'd have to review what the enemies are like, because if you're against a standing army you're going to have to slog through numerous turns on the opponents and you'd get caught out eventually by an exposed unit by sheer numbers. If you have full party-vs-party turns, they can lead to some broken strategies as mentioned.
I've personally had a tactical RPG in my head forever, and my thought would be stat-based initiative with a bar saying the next 3 or so units queued up and likely a limit to two enemy unit movements in a row. I also have the idea in my head that this game would be a 90's styled RPG with some obnoxious mechanics like time limits on turns, so... Take that as you will? :^)
I dont know about any of those examples you cited, but I'd be inclined to use a DnD framework, but you re-roll the initiative each turn and use modifiers based on skill or condition. If you have high adrenaline or have proficiency with a weapon, your initiative is improved (like fast reaction time or knocking an arrow faster). If you are blinded or dazed your initiative is modified down. The modifier could be a multiplier (rounded up or down), a added/subtracted integer, or a different number/type of dice. All this together (while making more math) should loosen up the play order without totally breaking the general capabilities across a given encounter.
I actually really like points based systems. Hoshigami, for instance, gives you a set number of points that you can spend on movement or attacking, and that gives you the ability to either move and attack like you can in most games, or if you don't have to move much, you can do multiple attacks. Alternatively you could end the unit's turn without using them and then their turn will come up sooner in the queue.
But of course, my problem with that is that if I were to implement that in the game, it would completely change how you play the game! Ah, that moment when you realize that you want to be making an entirely different game... ^_^;
Having a degree of random in turn order would just kill me though. In games where you're commanding a battalion and RNGesus has determined the unit you're depending on is just not going to go this round would drive me up the wall.
(Also just in case I weren't clear, by "tactics game" I meant games like Final Fantasy Tactics, Advance Wars, Nectaris, XCOM, etc. They're more commonly known as strategy games, but it's not a very accurate moniker.)
I've played a few games where fast units/characters get more turns, and I think it can be fine. In any turn based system, I hate nondescript loading bars with no clear markers slowly filling up for turn order (Bravely Default 2 does this). I've enjoyed games where a really fast character appears more often in the turn order, like ABC-ADF-ABC, as long as I can see the turn order so I know which units will move next (Stella Glow does this).
It might be giving too much away, but a game like Grandia with it's battle order bar made the turn based experience super legible. .
That said, it made it a little too obvious. I sort of missed the whole fog of war approach.
I think it can be fine to compromise with a generic "Enemy" turn marker to keep some element of surprise. Depends on what the developer wants to prioritize.
I just started replaying FFTA myself.
Personally, I like the "faster characters move more often" style of turn order, but like someone else said, the key to making this feel fair in tactics games is being able to see the turn order in advance, so you'll know when that thief is gonna get two turns to the paladin's single attack. Of course, characters can still affect their turn order by, eg. casting haste, or skipping one turn to make the next turn come faster.
One different way to do that is the way a lot of roguelikes do it, which is that speed is a number that goes down instead of up; that is, lower number is better, because the number indicates how many 'ticks' of the game clock a turn takes. This can be more comprehensible for experienced players; if the PC takes 5 ticks to move and the orc bouncer takes 8, you know you'll get an extra attack in every other turn or so and plan accordingly.
Finally, another system I like is where turn order is hard-locked in place, but the number of actions per turn, or the move distance, or damage or accuracy of attacks, is increased based on character speed. So, say a fencer and a full-plate greatswordsman each get one turn per round, but the fencer can move twice as far and her hits get a (speed – enemy.speed)% boost to chance-to-hit. This can result in the most tedious kind of kiting for fast characters and the most annoying kind of cant-hit-itis for slow characters, though, if not handled well; eg. the slow characters need either some way to pen fast characters in eventually or have special attacks that can inflict status or act at range.
The CT system in Final Fantasy Tactics for turn order is really transparent. When a unit hits 100, it’s their turn, any amount over 100 is discarded. There’s some funky logic for handling tie breakers (top leftmost unit goes).
Also the wait action gives 20 or 40 points to CT depending on if the unit moved or took an action on a given turn. So waiting has a tactical advantage.
The flaws in FFT are around high speed units. They gain a massive advantage at higher levels. There are two easy fixes. Flattening speed across the board so units like Ninjas, Archers and Thieves have a slightly higher speed stat, and slower units have a slightly slower stat. Remaining units have the same speed stat.
The other fix is that speed does not increase with leveling, as this makes balancing the game needlessly difficult. Spells in FFT do not scale in speed, and as a result spell casting really falls off late game in effectiveness.
Its really hard to talk about turn order in isolation. Because it's not just tied to other systems (action economy, turns-to-kill, stats) but also the fantasy you're trying to provide. And that's a matter of nailing down your project intention and subsequent design goals.
If you're working towards a semi-semetrical competitive PvP, your duty is to create a level field for creative expression where players need to create and block failure spirals. But for a narrative PvE, my philosophy is that you are working towards the illusion of balance but are really giving players license to game the hell out of the sandbox.
Like the Goblin Camp or Act 2 Tower Fights in BG3. For any semi-decent player, those fights are pure power-fantasy blood baths and some of my favorite. The turn order has nothing to do with it. Because even if the enemy has a dozen consecutive turns and overwhelming numbers, I can: overlap insane multi-turn AoEs, place units in the perfect ambush position and inject them in battle when I like, litter the field with traps, bribe enemies to my side and a host of other things that the enemy can't.
May seem off topic, but the it comes right back to turn order in isolation. If I had to go into those fights with a perfectly fair turn system and rational enemy AI, it would hardly be as impactful. There's no perfect system, just the right system built for the experience.
And another angle I like to explore is using game mechanics to inform the narrative. Like I had a tabletop game where 2 players were dating and couldn't help their characters doing the same. So I have a pair if players that are in sync in and out of game. They're constantly negotiating their insane medic/spy synergies. Other players are on their side. I said screw it, they can share turn resources and act in tandam.
Of course we balanced it with a ton of nerfs if they were solo or the relationship was compromised (The Therepy Tournament was a glorious shitshow). But its one thing for players to win a game. This was an event significant enough to change the rules. And now they don't just have new toys to play with, but they are also looking for other hidden ways to enhance their experience.
I personally like turn order being completely deterministic in tactics games. Whether it is by class (aka Tactics Ogre with their RT state) or by a speed stat of some sort that directly or at least heavily influenced turn order (as long as it doesn't end up being so good that it is the main stat that matters). I also like the more recent trend of being able to view the turn order and being able to manipulate turn order through the use of some abilities/skills (See: Tactics Ogre, Bravely Default, Octopath Traveler, Romancing Saga). The recent Romancing Saga RPGs are basically a masterclass in that where turn manipulation is a key part of the game.
Most recent tactics/turn based game in memory with a different turn order is For The King (1 and 2), where it's a combination of stats and status that determine turn order and as the battle progresses that turn order can change based on stat, status, and ability changes. It also clearly displays who is next in order to act along with highlighting the target in the turn order display when you select the target and allowing you to select the target from the turn order display.