[Release 0.1.0+] BATTLE MECHANICS

[Djinn_in_Tonic]

So far, our battle mechanics are extraordinarily simple: select ability, deal damage.

I know we're not really able to push out huge changes for our closest release date, but I'd like to debate things we COULD add to mix things up without adding too much complexity to the code base. I'll start us off with a few ideas.

• Poison status effect, to both demonstrate status conditions and since it suits the enemies in question.
• "Wounded" mechanic where abilities have a greater or lesser effect if the target is above/below 50% of current maximum hit points.
• "Flinching" / "Stunning" mechanic, where actors are pushed back in the action order (Flinching) or lose their upcoming action (Stunning).

[Roots]

I think the biggest immediate issue we face regarding our battles is that every enemy feels the same and the current set of skills we have for a character are more or less a decision between "deal less damage" or "deal more damage". Changing that I feel should be our immediate focus. We need to diversify the skills for both characters and enemies. For characters: every skill should have a situation where it executing it is "a good choice". For enemies, most of them (except maybe the slime) should have at least two skills to choose from, and the difference between those skills should be an obvious effect (ie one attack that deals large damage with a long warm up time, another weaker attack that often poisons the target). This is something that we can, and should, begin working on right now and we are capable of implementing various types of skills either immediately or with just a little extra work.

I'd like to debate things we COULD add to mix things up without adding too much complexity to the code base.

I'm not too concerned about code complexity. What I'm really worried about lately is play complexity. We already have a fairly elaborate battle system with a lot of features that are not trivial to comprehend. Multiple attack targets, degrees of intensity for status effects, "canceling" enemy actions, and battle fatigue. Adding in even more mechanics and making the battle system more complex isn't necessarily going to make it better. If a player is confused because the battle system is too complex and they can't "figure it out" quickly, they're likely to feel frustrated or simply stop playing the game. Having said that:

* Poison status effect, to both demonstrate status conditions and since it suits the enemies in question.
* "Wounded" mechanic where abilities have a greater or lesser effect if the target is above/below 50% of current maximum hit points.
* "Flinching" / "Stunning" mechanic, where actors are pushed back in the action order (Flinching) or lose their upcoming action (Stunning).[/list]

1) Poison status definitely agree we should add it. It's on my short list of features to add in the next week or two and I've already started looking into how to get it working.
2) Intriguing idea, but I don't know if that would make battles more fun if players always feel like they need to keep their characters healed above 50% to get maximum effectiveness out of their actions. Plus we'd need some way to convey that an actor is wounded (alternate idea: what if wounded was a type of status effect instead of dependent on HP amount?)
3) Is this the same as the cancel action mechanic I talked about from Grandia II in another thread? If so, we already do the flinching part, but don't have anything to do stunning yet (will take some code support to implement).

[Djinn_in_Tonic]

2) Intriguing idea, but I don't know if that would make battles more fun if players always feel like they need to keep their characters healed above 50% to get maximum effectiveness out of their actions. Plus we'd need some way to convey that an actor is wounded (alternate idea: what if wounded was a type of status effect instead of dependent on HP amount?)

The trick would be to mix-and-match. D&D 4e does this really well: some effects proc or have bonus effects when you're wounded, and some abilities deal more damage to wounded enemies. Some enemies, likewise, gain bonuses or penalties depending on their status.

It actually makes the fights pretty interesting, and gives a potential reason to heal up BEFORE you're at 12 hit points....which might be nice.

We also have a way to convey that they're wounded: the UI design I suggested turns health to red at below 50% Current Maximum.

[Roots]

Yeah, I figured changing the HP text/bar color could convey being wounded. But turning it red at below 50% feels like a bad design. Players would expect HP to turn red when it is at a critical state, so putting it at 50% would just feel pre-mature to me. But we don't have a clear way to convey that an enemy is wounded. And having wildly different things happen to an actor or skill while in the wounded state feels like it's too much information that the player needs to keep track of. I think it's a cool concept of having different changes happen while in a wounded state (some bad and some good), but I don't want to get crazy with it. Again, I'm really concerned about the complexity of our battle system and that we would be alienating all but the most hard-core strategist players. I want the game to be fun for casual players as well.

[Djinn_in_Tonic]

Yeah, I figured changing the HP text/bar color could convey being wounded. But turning it red at below 50% feels like a bad design. Players would expect HP to turn red when it is at a critical state, so putting it at 50% would just feel pre-mature to me.

Well, there's a reason it's orange-ish instead of red.

But we don't have a clear way to convey that an enemy is wounded.

Isn't that the goal of having enemy damage visibly displayed?

I think it's a cool concept of having different changes happen while in a wounded state (some bad and some good), but I don't want to get crazy with it.

Definitely avoiding getting crazy. I was more thinking of things like Healing spells more effective on Wounded targets (or possible abilities that have other effects and heal targets up to their Wounded value -- i.e. restore health to 50% if below), or Execute moves that deal bonus damage against Wounded foes.

Again, I'm really concerned about the complexity of our battle system and that we would be alienating all but the most hard-core strategist players. I want the game to be fun for casual players as well.

I'm not hugely attached to the Wounded mechanic, but I DO think it's a lot simpler than, say, our turn order combined with moves that interact with the flow of the turn order.

[Roots]

Isn't that the goal of having enemy damage visibly displayed?

Yes, but it was meant as a visual reference and not as a clear means of discerning the state of an enemy. Remember there are four different damage frames, and it's not always apparent which frame is which (some are very close together, others are a less-obvious way of conveying damage, such as the fire tree enemy whose flames get weaker and die out). Which one(s) of those would correspond to the wounded state? That's really what I was referring to when I said that we need to convey that the actor is in a wounded state.

Definitely avoiding getting crazy. I was more thinking of things like Healing spells more effective on Wounded targets (or possible abilities that have other effects and heal targets up to their Wounded value -- i.e. restore health to 50% if below), or Execute moves that deal bonus damage against Wounded foes.
I'm not hugely attached to the Wounded mechanic, but I DO think it's a lot simpler than, say, our turn order combined with moves that interact with the flow of the turn order.

That sounds interesting. I don't dislike the wounded mechanic by any means, and what you said sounds intriguing if we could find the right use of this mechanic and make it easy to convey when an enemy is in a wounded state. I'm just worried that we're discussing adding even more features onto an already full plate and that it would make a player even more exasperated at all the complexities of our system. However, if we were removing another feature and replacing it with this one, I would be much less apprehensive about it. I've actually been pondering about the multiple attack targets feature and wondering if that really makes the game more fun, or if it just makes it more tedious and overly complex (just thinking at this point though, not planning to propose we remove it).

[Roots]

So on another note, I want to discuss a little more about status effects and elemental effects as well.


Elemental Effects
These are a familiar mechanic in RPGs, and our old designs had plans for this as well. Fire-based enemies are weak to water attacks, for example. Normally elemental effects only applied to magic-based attacks, but in Allacrost we have physical-based elementals as well as ethereal (magical) elementals. There are four of each.

Physical: Slashing, Piercing, Mauling, Smashing
Ethereal: Fire, Water, Volt, Earth

An enemy may be weak to or strong against any of these eight elements. Our idea was to have a sort of paired relationship for the ethereal types that would be familiar to RPG players: enemies strong against fire are weak against water, for instance. So we have a cyclic chain of "strong against/weak against" that looks like this: "Water -> Fire -> Earth -> Volt ->". We don't have this sort of relationship for the physical types, mostly because it wouldn't be intuitive. Instead, we just say that an enemy may be weak to both piercing and smashing, but strong against slashing, for instance. We also plan to have the same levels of intensity for elemental effects that we do for status effects (essentially, a higher intensity just increases the resistance or weakness to that element).

We've purposely put elemental effects on hold and didn't plan to begin introducing them until the next chapter of the game (to keep the amount of learning material in the first chapter down). We can use these properties to make battles a little more interesting, and make certain characters/abilities better against specific types of enemies. That's all I really want to say about this for now. This is just a FYI as this is in the game plan.


Status Effects
I want to brainstorm some status effects both for the future (next year), and what we could begin using immediately in our releases this year. Right now we have available the following:

• Poison (currently being developed)
• Stun (enemy doesn't move on action bar until it passes)
• Attribute modifiers (temporary increase or decrease to a single attribute for an actor, like strength or agility)

One thing I want to avoid is having multiple status effects that are nearly identical. For example, many RPGs have Sleep, Stone, and Paralyze which do basically the same thing (prevent the character from taking actions) and only differ in the means in which you heal them. I also want status effects to all be useful. In other words, if we give access to a skill for a character that applies a status effect and the player never finds a good use for it, we failed with this effect.

I also want to address the current attribute modifier stats. We have a lot of different attributes (strength, vigor, fortitude, protection, stamina, resilience, agility, evade). That's 8 right there, and would be 10 if we add HP and SP to this list as well. Having a status effect for each attribute feels like a bit much, so I'm wondering if we can combine similar stats into a single effect to make the list a little more manageable. Here's a proposal:

• Modify attack power (changes both strength and vigor by the same amount, or apply a multiplier to the normal damage dealt for attacks)
• Modify defense power (changes both fortitude and protection, similar to above)
• Modify fatigue generation rate (changes both stamina and resilience, similar to above)
• Haste/Slow (basically the same as modifying agility stat, maybe this could apply to evade as well?)

Note that some status effects are "opposites". In other words, you can't have both a Haste and a Slow status active at the same time. Casting a Haste spell on a character that has Slow active reduces the intensity of Slow and may cause the character to have a less intense haste effect. For example, let's say we have a powerful haste skill that applies a 3rd degree intensity to the target. A character currently has a 2nd degree intensity Slow status active. Casting this haste skill would remove Slow and apply a Haste status of 1st degree intensity (because Haste^3 - Slow^2 = Haste^1). Make sense?

Finally, another issue to consider is how to determine how long a status remains active for (recall that status effects gradually reduce in intensity over time until they dissipate completely). At the moment, status effects have a customizable duration in real time. For example, the stun effect could have a duration of 2 seconds per intensity level (ie 4th degree intensity stuns for 8 seconds total until the status is completely cured). A poison status may stay active for 10 seconds per intensity level. Should we continue to use a real-time based approach, or instead use a turn-based approach to status effects? In other words, poison remains active for 3 turns for a character. (Stun wouldn't work with a turn-based approach though, since it literally prevents the actor from taking turns).


I know that's a lot to take in. I think that the current mechanics for status effects are okay, but wanted to throw the whole design on the table in case there are issues we need to consider. For now, I'd like to toss around ideas of different status effects we could have that would make battles more interesting and strategic without also being an annoyance to deal with. I'll throw out some very quick ideas that I haven't thought much about to get things started.

• Stasis: actor is not able to be targeted or take any action
• Nullify Fatigue: temporarily removes some or all fatigue effects, allowing HP/SP to be restored to a higher max for a period of time
• Confusion: characters select random action and players can not control them (unsure what this status effect would do for enemies though)
• Exhaustion: Skills require more SP to use while active (which likewise produces more fatigue)
• Vendetta: attack power increases and defense power decreases by the same degree

[Roots]

The poison status effect is just about ready for use. I need to do some testing to make sure it works correctly, but the framework is all there. I currently have the effect lasting for 20 seconds total (per intensity level) and damage is dealt from poison every 8 seconds. The amount dealt each time is equal to 0.5%, 2%, 4%, or 7.5% of the actor's max HP (the full max, not the current max when fatigue is applied). The amounts vary according to the effect's current intensity.

All these numbers are easy to change and only require modification in the Lua file (lua/data/effects/status.lua). I think these numbers are reasonable for a first attempt, and we can refine and rebalance them as needed once we get an idea for how powerful the effect is in the game. I'll be adding this effect to the spider's and scorpion's list of attacks.

[Roots]

Poison status effect and status effect code redesign changes are now live in the repository. Everything is working correctly, although there's still some bumps to roll out. For example, a "hit" with a poison effect can sometimes miss, because it's being treated as a type of attack. This should be pretty straightforward to fix.

[Roots]

Do we want to have skills that consume no SP?

Up until now we've always had at least one skill that didn't consume any SP, because the character needs to be able to do something every turn. Does it make sense to have a generic "Recover" action that all characters can select, which will essentially idle the character for a turn and allow them to accumulate SP? Or do we prefer having at least one 0 SP skill for each character? Since SP regenerates naturally over time, we can have their 0 SP skill do wildly different things: a very low damage attack, defend for a turn, or maybe regenerate a little amount of health.

I can see pros and cons to both options. Having a common Recover action for a character provides for some consistency. Having 0SP skills provides for more flexibility and uniqueness. I'm kind of leaning toward having a Recover action, just because I don't think we gain much from that flexibility with having 0SP skills (the individual character skills themselves should provide plenty of flexibility and uniqueness for characters). Opinions?

[Djinn_in_Tonic]

I can see pros and cons to both options. Having a common Recover action for a character provides for some consistency. Having 0SP skills provides for more flexibility and uniqueness. I'm kind of leaning toward having a Recover action, just because I don't think we gain much from that flexibility with having 0SP skills (the individual character skills themselves should provide plenty of flexibility and uniqueness for characters). Opinions?

I think the Recover action is the way to go here, for both clarity and consistency. I'd even be okay with putting some sort of penalty (takes additional % damage while recovering based on how much SP is missing when recovering, for example) to encourage the player to recover strategically rather than when at minimum SP, but that might be difficult to alert the player too...

Thoughts on that? I think it's actually a pretty elegant mechanic to encourage players to think ahead and avoid simply burning through all their SP on powerful abilities, but the variable nature of it may be disconcerting to players. I *think* it's a problem worth solving, but I'd like opinions.

[Roots]

Okay, sounds like we are in agreement here so let's go with a design where all skills consume at least 1 SP and have a recover option. Now that I think about it, because a character recovers a small amount of SP every turn, they should never be at 0 SP when they are selecting an action. Maybe the recovery should simply amplify the SP recovery amount by 3x or 4x what it normally would be?

I'm okay with the idea of implementing a penalty while the character is in a recovery state. I'm feeling about a dynamic penalty though. It would increase strategic decision-making, but it would be difficult to convey this information to the player so they could make an informed decision. Providing a static penalty is much simpler to convey, for example increasing damage taken by 40%.

If we want to get a little more fancy with it, maybe we could have a small number of types of recovery that the player can choose from. For example:
- Basic Recovery: increase damage taken by 30%, increases SP regeneration by 3x
- Long Recovery: increase damage taken by 40%, increase SP regeneration by 4x, increases time in recovery state by 2x
- Dual Recovery: increase damage by 25%, increases SP regeneration by 2x, and also restores HP by the same amount


But this might make the recovery option needlessly complicated. For the immediate future (this month's release) I suggest we do the following:

• Add a Recover option in the action type menu
• Recover generates the normal SP regeneration by 150%
• Takes a very short warm-up time (500ms)
• No penalty applied for using (you're already "implicitly" penalized by consuming a turn without attacking or using skills)

Thoughts on this immediate plan? I'm more concerned about what we want to change now than what we want this option to do in the future. We can always further develop and modify what this action does in later releases.

[Djinn_in_Tonic]

But this might make the recovery option needlessly complicated. For the immediate future (this month's release) I suggest we do the following:

• Add a Recover option in the action type menu
• Recover generates the normal SP regeneration by 150%
• Takes a very short warm-up time (500ms)
• No penalty applied for using (you're already "implicitly" penalized by consuming a turn without attacking or using skills)

Thoughts on this immediate plan? I'm more concerned about what we want to change now than what we want this option to do in the future. We can always further develop and modify what this action does in later releases.

I'd make it a separate heading, actually. Skill / Recover / Item. It's not really a skill, after all. Regardless, I'm okay with this as an immediate change -- I think we can build an interesting SP system around this.

[Roots]

I'd make it a separate heading, actually. Skill / Recover / Item. It's not really a skill, after all.

Yes, this is what I meant when I said "action type". Skill/Recover/Item is all a type of action. I'll add a basic Recover action this week that is good enough for the release. We'll see if we want to make this action a little more interesting at a later date. For now, I want to continue focusing on fatigue and skill balancing.