Monday, January 22, 2024
Stat Overhaul Details!
However, as feedback showed, these steps were not a strong enough foundation to fully establish STR, DEX, and INT as sufficiently distinct stats. Further, the old and incomplete paradigm for accuracy remained in place for pets and guests, needing the secondary stat phase to be completed before being implemented. In order to have time to properly distinguish the main stats, tackle this accuracy issue, and address secondary stats as well, we needed to set a release aside to work on everything in one go. Unfortunately, we hit a major obstacle... old files.
Any changes to Charisma, particularly the guest changes detailed below, would require affecting guests in some manner, but guest costs and stats have always been handled by their individual files, not by the game engine. Before we could continue, Ianthe had to manually adjust hundreds of files (revamping many in the process) to incorporate them into new code that allows the engine to dynamically manage guest parameters. That effort required many small releases on our part, and was disrupted by health matters, but it is now complete.
With the preliminary work now out of the way, we're now prepared to announce the complete stat revamp release and explain the upcoming changes in full!
Due to the critical nature of the stat overhaul, it has been given top priority and is set to commence at the earliest availability on the calendar in February. This will however require rescheduling the cross-game GGBs to April. We appreciate all of your continued suggestions in the ongoing thread, but kindly limit your suggestions to 3.
The following is a summary of the changes to each upcoming stat:
Base Behavior Changes:
Enable "Warrior Lean": FD armours deal 100% outdoing damage and take 80% incoming damage.
+15% to all weapon damage. First weapon attack of a battle inflicts Choke (-12.5% damage/4 turns) with no save and a 10 turn cooldown. We MAY change this to require a Melee attack if we find that it tips STR toward only being taken for the bonus and not for melee attacks.
Base Behavior Changes:
Ranged damage starts at 80%. Whether or not you're using a Ranged weapon on a given turn, it increases by +7% per turn until it hits 101%, then +3.5% per turn until turn 20.
Enable BTH lean expertise: the downside of BTH leans is mitigated by up to 25% on all weapon attacks and spells.
>> Example: at +20 BTH lean, you normally gain +20 BTH and deal 81% damage. With full DEX, this becomes +20 BTH and deal 86% damage.
>> Example: at -20 BTH lean, you normally deal 131% damage and take -20 BTH. With full DEX, this becomes 131% damage and take -15 BTH.
Base Behavior Changes: None.
Wallbreaker: Spells pierce enemy elemental resistances and gain +[(INT/ExpectedINT)*(100-MonsterEleMod)/200]%
>> Example: Against 90% resistance, spells deal +5% damage. Equivalent of hitting 95% resistance.
>> Example: Against 50% resistance, spells deal +25% damage. Equivalent of hitting 62.5% resistance.
>> Example: Against 10% resistnace, spells deal +45% damage. Equivalent of hitting 14.5% resistance.
Base Behavior Changes:
ENEMY ONLY: Reduced health scaling.
[Implemented] Increased status resistance [Implemented] Increased heal resistance
Base Behavior Changes:
Pet and guest accuracy is now based on [Mainstat+CHA] instead of DEX+CHA.
Guest stats adjusted to account for paying a third of the actual cost of their output: Base guest output reduced to 45% Melee from 60% Base guest Cost increased to 30% Melee from 21%
Guests gain Ferocious Strikes: A flat 22.2% chance to have double output (Damage, status effects, boosting features, etc).
Note: After style bonus, average guest output becomes 55% melee*
Base Behavior Changes:
Removed: Bonuses to BTH and blocking Added: Global 5% chance of hitting, regardless of defence. Player-only: with 250 LUK, this can increase to 10%.
20% chance per turn to flip one status debuff on the player into a status buff.
Warrior Lean is a community suggestion that we thought was perfect for providing a defensive Warrior identity without subsuming the FD Ranger niche. To answer a frequent question, this would not be the addition of a new lean. Instead, all FD armors would lose their player damage penalty when wielded with Melee weapons. Warriors would then deal normal damage with attacks, while continuing to take reduced damage.
The originally proposed style bonus of a small boost to all weapon damage was underwhelming, so we're scaling it up to +15% to all weapon damage. This should play into STR's core niche while making the stat appealing to hybridize into while getting to have a mix of weapons. The remaining 5% Melee value is spent on a guaranteed choke on the first successful attack in order to further solidify the Warrior niche of the steady attacker through long bouts without stepping into END's role.
The original iteration of a ramping damage identity was initially negatively received, but testing produced a much more favorable reception once it didn't take so long to achieve parity. The finished version will still ramp up quickly, giving it a strong setup-and-blast identity that could be leveraged for both offensive and defensive builds. Final values cap out at turn 20 so as to keep from outperforming STR at its own niche -- It would defeat the point of the revamp to reverse the problematic linking of the two stats instead of actually fixing it by giving them two distinct roles.
The style bonus allows DEX to play with a wider range of damage leans, quickly overturning the disadvantages of accurate weapons or doubling down on raw damage with inaccurate ones. Part of it is dedicated to blocking, making DEX more of a stat that gives a headstart to blocking as opposed to the blocking stat.
Due to its place as the sole stat with a functional identity, INT will receive no changes to its base behavior. INT is at the level quality we want other stats to be at, and very much didn't need any vertical power growth, so its style bonus is focused on versatility. Low resistances have a much harder time denying elemental damage from spells, making those pesky tank mobs less of an obstacle to a caster with a diverse toolkit.
Although it's not as hard of a rule as with STR, taking away from END's HP value to dedicate it to something else would still require extensive rebalances and risk flipping the stat into an underpowered status. Therefore, this is a difficult stat to make interesting, though we are of course open to feedback on the matter. However, player feedback and low level testing made it clear that this stat scales way too aggressively at lower levels, so the prior step of this revamp featured a separation of health scaling to address this. If there are any similar problems with other stats, please let us now in the replies!
Style bonuses remain the same as the prior iteration.
Guests have always provided several times the power they paid for, as they were originally designed based on a damage and accuracy model that was never implemented. Their cost discount was meant to be a decompression bonus for using Beastmaster weapons, as part of this model in which CHA would have been a main stat and DEX a secondary one. Since this old model was never completed, and guest costs couldn't be dynamically adjusted, the end result is essentially a spell being cast every turn that only pays part of its actual cost. That really isn't fair to any other build.
However, reducing guest power to what is being paid for would feel awful, and just increasing the cost would be just as harsh. The adjustment needed to be split, with a further way to mitigate the resource cost increase. We sourced the above value adjustment from mixed community feedback in June 2023, and the Style Bonus is fully dedicated to guest Ferocious Strikes, so that keeping costs manageable can be part of the stat behavior itself.
Instead of having a mandatory role in accuracy and blocking, LUK now has a unique role in a new accuracy system. Attacks always having a hit chance floor should address the frequently reported issue that blocking armors are a one-and-done deal. It was widely recognized that blocking builds could make the game uninteresting by only getting hit by rare autohit attacks, making them only really need one armor. Now these builds can be catered to with a full spread of evasive equipment, and LUK allows players to double their own hit chance floor to increase reliability against dodgy mobs.
The style bonus had to be something entirely new so as to not make any one function of a generalist stat overtake others, while not further powercreeping lucky strikes in a situation where they can be easily guaranteed. The opposite idea practically wrote itself. Now LUK will give you both lucky strikes and lucky breaks!
For the entire run of the prior stat paradigm, the only main stat that could be said to have an identity of its own was INT through paying weapon damage in exchange for mana to frontload damage with. STR builds had the vague identity of a steady damage dealer, but gear support and booster disparity made INT builds better at STR's own niche. Any gear meant for all builds, even if Warrior-oriented, was better at dealing steady in the hands of Mages. Warriors had no interesting baseline stat functionality, as the fact that STR dedicates all of its power to damage made it the root of the game's mathematical valuations.
As of Rangers: Even while being the primary stat for Ranged damage, DEX was still dependent on Strength for its damage component, while pulling double duty as a requisite secondary stat for Melee and Magic accuracy. Even pet attacks required it for accuracy (They no longer will as of the next update).
Consequently, offensive Rangers functionally couldn't have a niche of their own. They required the same stat spreads as warriors, and the accuracy leans of Ranged weapons meant that STR+DEX builds were generally better off choosing melee weapons. Defensive Rangers had a small niche of their own in a focus on resource efficiency over time efficiency, but this could still be infringed on by defensive spellcasting builds and to this day needs fleshed out gear support.
In short, main stats were in such a state of crisis that only INT could receive truly unique gear support, STR's identity was being subsumed into INT, and DEX's identity was nonexistent outside of what small niche couldn't be subsumed into STR.
Secondary stats had a schism of their own:
END being mathematically overpowered but exceedingly unpopular even against lengthier boss fights made for an understandable but nonetheless frustrating situation. The stat did one visible thing, adding health. This one thing would have to be weakened if some of its power were to shift towards doing something more interesting, at which point it could easily turn very underpowered. This intersected with the DEX situation, as END could have at least been recommendable for new and returning players if not for the stat point tax that DEX represented. Most egregiously of all, END scaling on enemies was exceptionally punishing for lower level characters.
LUK has a less generic role due to being a mix of the crit stat and the "Jack of all trades" stat, but doing a lot of things at once meant that it was difficult to buff any one of these things, leaving it largely as the former. Modern issues with it are largely outside of the stat revamp proper, as making critical hits more interesting would require a separate status effects and critical hit project.
CHA was coded in a way that was a vestige of another, completely different stat plan that never came to pass. Combined with old file issues, this led to both being tied to DEX and guests getting a cost discount in exchange for a weapon requirement that was not even possible to apply at the time.
Any one of these issues, on its own, would severely hinder our ability to provide cool rewards for a build. The interwoven combination of them was therefore a huge call to come up with a way to leave every single build at a stage where it could stand on its own and was ready to be fleshed out in the long term by build support. To that end, the revamp includes a new addition to every stat:
While we will need a separate tutorial revamp to have adequate in-game introductions to the game's mechanics, it's not enough to simply tell players what each stat's purpose is. Especially if it seems like their gear encourages a different style. Each stat itself should reward playing by its style by being fun and powerful in its own unique way. However, as mentioned above, STR and END could only do one thing, since STR can't shift any of its power and END could easily become underpowered if it did.
To that end, we're implementing the Style Bonus: A small budget of 20% Melee in power, added to every stat in the shape of small boosts that reward its core behavior or shore up its transitional issues. In the case of main stats, these bonuses are meant to reward both pure and hybrid builds by making each stat equally appealing as a primary or additional main stat. The Style Bonus is unlocked upon reaching 150 in a stat, beginning at a quarter of its power, and scales up to its full value with every 5 points until reaching its full 20% melee value at 250.
The Style Bonus is a major boost to the power of each stat, and thus constitutes quite a jolt of power creep. However, this is a calculated addition meant to be specifically leveraged towards making each stat more fun to play in its intended niche. Because of this, the Style Bonus is a player-only feature. These advantages are based on player playstyles, and not meant for mobs. We recognize that it's not always fair for enemies to play by the exact same rules as the players, and are exploring other divergences that could make things more fair and fun without being too hard or too easy.
Thorough as this may be, this revamp is still a work in progress! If you have any feedback or general ideas, please let us know! Is there anything we missed? Are there other stats in need of an enemy scaling adjustment? Any wild ideas that you think would improve the game?
If you'd like to weigh in on the balance of the changes, be sure to familiarize yourself with this explanation of what balance means and how we apply it as essential reading before posting. Make sure to post in a friendly and constructive manner as per the forum rules as well -- Moderation will be stricter with any hostility in this thread in order to be able to keep it open, as we really do want your feedback.
As a final note, we do realize that stat changes alone, no matter how massive, can't solve all early game hurdles by themselves. Old quests that predate scaling and issues specific to mobs are a matter for future projects and announcements, but we are absolutely exploring solutions!