Spellsword (formerly Addon reworking Arcane Blade Class)

[Hirumakai]

First, thanks for the replies and feedback.

I had not realized staves were accuracy weapons nor realize what their bonus was. Thanks for pointing it out. This is what I get for making a major change with only a single day of testing.

So in combat.lua in the AttackTargetWith function we have:

if self:isAccuracyEffect(weapon, "staff") then
local bonus = 1 + self:getAccuracyEffect(weapon, atk, def, 0.025, 2)
print("[ATTACK] staff accuracy bonus", atk, def, "=", bonus)
self.__global_accuracy_damage_bonus = bonus
end

This is after damage from a swing is applied, but before the extra bonus damage gets applied, i.e. the on melee hit does +5 fire damage and the like effects/procs. For example lightning melee bonus damage from Air stance. Or a spellsword proc.

The global accuracy damage bonus is set back to nil at the very end of that function as well.

In Default Projector we have:

if src.__global_accuracy_damage_bonus then
dam = dam * src.__global_accuracy_damage_bonus
print("[PROJECTOR] after staff accuracy damage bonus", dam)
end

So the number being passed is a straight up damage multiplier, which normally goes between 1 (at equal to or less accuracy than defense), to 3, when accuracy > def by 80. So master of weapons would increase that at high end to something like a 7.7 multiplier (with the 1.2 category multplier its more like 285% instead of 250%). Yeah, that is some kind of crazy. As you say, the base staff is strong as well with up to a x3 multiplier.

Spellsword combat procs do damage based on

getDamage = function(self,t,eff_,talent_level) return t.modifiedCombatTalentSpellDamage(self,t,1 + eff_talent_level, 25, 225) end,

Where modifiedCombatTalentSpellDamage is based on 1 + the average effective talent level of the windcutter tree talents (i.e. Air Strike, Tornado Strike, Windstorm and Whirlwind Teleport).

So with a 1.3 multiplier and 5/5s in all talents, and 100 spellpower, we're looking at a base proc damage of 270 damage. With a base staff and assuming 80 accuracy, that could go up to 800. Master of weapons could push that as high as 2000 per proc. AOE.

Plus air stance provides it own on hit damage proc of about 50 lightning damage at 100 spellpower. That could go as high as 150 with a staff and 385 with master of weapons.

Yeah that wasn't intended.

The intention was for martial mastery to be a melee non-magical type tree. Something not out of place on a Bulwark, Berserker or Rogue. I think the correct course of action is to exclude staves, and leave them at their default 0.025, capped at 2, scaling.

Any other effects looking like they might be unintended. Also, assuming master of weapons only applies to traditional weapons as opposed to staves, does it seem to be worth using?

I'll be pushing a fixed patch to it later today.

[Hirumakai]

Version 1.2.3a of Spellsword has been pushed.

It includes a fix where Master of Weapons no longer affects accuracy effects of the "staff" type, which includes shields as well as staves.

Not sure if the talent is worth it as written, but will continue to play test. Further comments are always welcome.

[malboro_urchin]

Version 1.2.3a of Spellsword has been pushed.

It includes a fix where Master of Weapons no longer affects accuracy effects of the "staff" type, which includes shields as well as staves.

Not sure if the talent is worth it as written, but will continue to play test. Further comments are always welcome.

Why should the talent not affect shields? You wanted the talent to fit with martial non-magical classes, and Bulwarks in particular rely on the shield as a good part of their offense, with their many shield bashes and whatnot,

[NemesisZeru]

Version 1.2.3a of Spellsword has been pushed.

It includes a fix where Master of Weapons no longer affects accuracy effects of the "staff" type, which includes shields as well as staves.

Not sure if the talent is worth it as written, but will continue to play test. Further comments are always welcome.

Why should the talent not affect shields? You wanted the talent to fit with martial non-magical classes, and Bulwarks in particular rely on the shield as a good part of their offense, with their many shield bashes and whatnot,

Considering Earth Stance uses shields(Earth Strike does a shield bash, as does Sand Blast), iunno. Seems weird.

From the looks of it, it's...

Fire: Twohanded weapon
Earth: Sword and Shield
Air and Water: Dual weapons?

[Hirumakai]

Why should the talent not affect shields? You wanted the talent to fit with martial non-magical classes, and Bulwarks in particular rely on the shield as a good part of their offense, with their many shield bashes and whatnot,

Considering Earth Stance uses shields(Earth Strike does a shield bash, as does Sand Blast), iunno. Seems weird.

From the looks of it, it's...

Fire: Twohanded weapon
Earth: Sword and Shield
Air and Water: Dual weapons?

I excluded the shields as well as staves because the shield accuracy effect is a 2.5% increase in spell proc damage, the same as staves. People would be spamming Earth strike and Sandblast to get something like x8 proc damage best case. Thats like a 2000 damage AoE.

Basically the staff and shield accuracy effects are way out of line with the rest of the weapon accuracy effects when applied to Arcane Blade and Spellsword style spell on hit procs. They also aren't thematic in whats intended to be a purely non-magical/powerful through training/incredible skill type of tree.

I'm not currently in love with the talent. It was something I cooked up fairly quickly to replace the old tier 1 talent in the tree, which was Weapon Bond, which turned a normal weapon into an artifact weapon. Its power grew as you killed more things with it (especially bosses). But coming back to it after a year, I didn't like how it changed the overall roguelike gameplay, as the best course of action was go shopping for a weapon with base traits you wanted, and then keep it for the entire game.

Still brainstorming for a possibly better, accuracy themed talent. Suggestions are welcome. I may also tweak this version some more on further testing.

[Doctornull]

Two fixes I can think of:
- Don't - with accuracy bonuses. They're not as terrible when they don't do as much.
- Change the accuracy bonuses of staves and shields. They'd be less terrible if they did something else.

[Hirumakai]

I'm thinking of replacing the first tier talent with the following:

Master of Combat:
Sustain (20 stamina):
As a master of combat, the user always does some damage. Each time the user attacks, irregardless if he hits or misses, he deals X physical damage. If the user does miss, they gain a bonus of +Y accuracy for Z turns. This stacks with itself and will refresh the duration.

Y = +2 accuracy per talent point
Z = talent points

The X damage is a 0 radius ball projection on a square after an attackTargetWith call, and ignores whether it hit or not. The damage is self:combatTalentAttackDamage(t, 5, 40), which is like combatTalentSpellDamage except uses attack (aka accuracy) instead of spellpower. So at 5/5 and 100 accuracy it would do 40 bonus physical damage which is spell like in that it always hits and ignores evade/deflect type abilities.

The damage scaling is identical to Fiery hands (other than the spellpower -> accuracy swap), and instead of generating stamina, it generates accuracy on misses. I dropped the equivalent mana-> stamina sustain cost as technique stamina sustains tend to be lower than spell mana sustains. The resource pool is smaller, since willpower gives 2.5 stamina, but 5.0 mana.

Thoughts?

Also, I found a bug in martial magic which is making talents not utilize resources once you have it. I hope to patch it and the 1st tier talent later today.

[Hirumakai]

I've pushed a 4th version (Spellswordv123d) with further bug fixes.

Earthen Shield I believe may had a bug which rarely put the game into an infinite error loop with getDisplayName errors. I have removed the custom display name for Earthen Shield as didn't seem to be updating/refreshing the hit points left in it anyways.

I also fixed a bug where it wasn't actually preventing damage. It would tick down at half the damage dealt to the character, but the character would still take full damage. This should improve the survivability of those using the talent.

[NemesisZeru]

I'm thinking of replacing the first tier talent with the following:

Master of Combat:
Sustain (20 stamina):
As a master of combat, the user always does some damage. Each time the user attacks, irregardless if he hits or misses, he deals X physical damage. If the user does miss, they gain a bonus of +Y accuracy for Z turns. This stacks with itself and will refresh the duration.

Y = +2 accuracy per talent point
Z = talent points

The X damage is a 0 radius ball projection on a square after an attackTargetWith call, and ignores whether it hit or not. The damage is self:combatTalentAttackDamage(t, 5, 40), which is like combatTalentSpellDamage except uses attack (aka accuracy) instead of spellpower. So at 5/5 and 100 accuracy it would do 40 bonus physical damage which is spell like in that it always hits and ignores evade/deflect type abilities.

The damage scaling is identical to Fiery hands (other than the spellpower -> accuracy swap), and instead of generating stamina, it generates accuracy on misses. I dropped the equivalent mana-> stamina sustain cost as technique stamina sustains tend to be lower than spell mana sustains. The resource pool is smaller, since willpower gives 2.5 stamina, but 5.0 mana.

Thoughts?

Also, I found a bug in martial magic which is making talents not utilize resources once you have it. I hope to patch it and the 1st tier talent later today.

Seems interesting enough to me. Might be a little weak, though..but then again, I don't know much about Arcane Blades, haha.

[Hirumakai]

I pushed a new version, Spellswordv124a. This was mostly a clean up pass, trying to get the high level scaling in line with current talents. Still haven't coded up different talents for martial mastery (although did actually make Master of Weapons do what it says, namely add an accuracy bonus).

Change log:

1) General clean up of code style

2) Modified Elemental strikes (Air,Earth,Fire,Water) to correctly limit their center to 1 square away, even without a target.
code:
if core.fov.distance(self.x, self.y, x, y) > 1 then return nil end

3) Modified Elemental strikes to use combatTalentLimit to limit high level stats to maximum duration of 3 and maximum radius of 9. This changes the duration/radius progression slightly at low talent points.

code:
getDuration = function(self,t) return math.floor(self:combatTalentLimit(t, 4, 1, 3)) end,
radius = function(self, t) return math.floor(self:combatTalentLimit(t,10,0.5,2.5)) end,

4) Inferno cleave had its heal mod changed to match Bleeding Edge.

code:
healloss = function(self,t) return self:combatTalentLimit(t, 100, 17, 50) end,

Also explained that the Str mod of the main hand weapon is used to multiply the flaming wound damage. I.e. normal damage with a 100% Str weapon (1-hander basically), and 120% with your typical 2-hander. A stave, not having a strength mod is only 0.6.

5) Aura of Fire capped at 100% chance.

code:
getChance = function(self,t) return math.min(math.floor(self:combatTalentSpellDamage(t,20,50)),100) end,

6) Water strike ice block hit points changed from:
local hp = 70 + self:combatPhysicalpower() * self:getTalentLevel(t)
to
local hp = 70 + t.getDamage(self,t) * 3
where
getDamage = function(self, t) return self:combatTalentPhysicalDamage(t,10,135) end,

This is more in line with the Freeze spell's ice block effects.

7) Healing waters had the allow_on_heal flag set for the intial heal or residual heals if the amount actually healed is 100 or more at once. Healing waters can now crit, but the maximum residual carry over is the base heal amount. This will allow it to scale better late game, but without people preping it multiple times waiting for a crit.

8) Tidal rush had its targeting type updated to beam, which is closer to what it does, given intervening enemies are either knocked out of the way or swapped with. It represents riding a giant tidal wave over all of them.

Pin duration set to a maximum of 3 turns.

Range changed to same scaling as rush, but with 2 less in min and max (i.e. 4-8 instead of 6-10):
range = function(self, t) return math.floor(self:combatTalentScale(t, 4, 8)) end,

9)Drain water of life now has allow_on_heal flag set when it does the heal. Also fixed an incorrect info block.

10) Fixed a typo in the damage Earth Strike main hand weapon was doing (1,1.5) scaling instead the correct (1,1.45) scaling.

11) Learning Earth strike should now mean shields show their melee damage properties when examined.

12) Modified blind duration on Sandblast to be
getDuration = function(self,t) return math.floor(self:combatTalentScale(t, 1, 5)) end,
instead of 1 turn per raw talent point

13) Modified Rolling earth radius and stun duration to match Flameshock

radius = function(self, t) return math.floor(self:combatTalentScale(t, 4, 8, 0.5, 0, 0, true)) end,
getStunDuration = function(self, t) return self:combatTalentScale(t, 3, 7, 0.5, 0, 0, true) end,

14) Changed Tornado strike buff duration to (still 3 max turns):
getDuration = function(self,t) return math.min(3,self:combatTalentScale(t,1,3)) end,

15) Changed Whirlwind teleport to the same range scaling as Tidal rush:
range = function(self, t) return math.floor(self:combatTalentScale(t, 4, 8)) end,

16) Buffed Arcane Stamina to be roughly 1/5th of the Quick Recovery passive per sustain, up from a flat 0.1 per point per sustain.

getStaminaMultiplier = function(self,t) return self:combatTalentScale(t, 0.1, 0.5, 0.75) end,

17) Dropped Mana Focus's stamina sustain cost to be mid-way between Arcane Feed and Shadow Feed. Mana regen was placed on the same regeneration scale as Arcane feed and Shadow feed (this is a slight increase).
Mana per hit remains the same as Shock Hands.

getManaRegen = function(self, t) return self:combatTalentScale(t, 1/7, 5/7, 0.75) end,
getManaPerHit = function(self,t) return self:getTalentLevel(t)/3 end,

18) Martial Magic duration was nerfed to be in line with Adrenaline Surge. So instead of 4,5,6,7,8, its now 3,4,5,6,7

getDuration = function(self, t) return math.floor(self:combatTalentLimit(t, 24, 3, 7)) end, -- Limit < 24

19)Spellsword combat uses the same chance scaling as Arcane combat, except when using a shield it is at 80% instead of 75% as there are fewer shield-bash type abilities the class has access to.

20) Arcane Prowess switch over to use combatTalentScale, going from 10 to 30 (where Arcane Cunning goes from 20 to 40).

code:
getSpellpower = function(self, t) return self:combatTalentScale(t, 10, 30, 0.75) end,

21) Updated the resistance penetration on Elemental Destruction to match that of Wildfire.
code:
getResistPenalty = function(self, t) return self:combatTalentLimit(t, 100, 17, 50) end, --Limit < 100%

22) Fixed a display bug on Earthen Shield's regen rate (its capped at 10%), the info display keeps going up.

23) Moved superloading onActBase to be callbacks instead for Earthen Shield, Wind Storm, and Aura of Fire.

24) Fixed Master of Weapons actually adding accuracy

[stinkstink]

Some feedback after playing down to ID150. Character's here if you want to take a look:

-I'm not a fan of Purifying Fire having a max duration, considering it requires a Spell Save check and the enemies I most want to hit with it (stairbosses using Unstoppable or Swap Places/Evasion) tend to have 120+ spell save due to getting 22 levels in Spell Shield. I usually run around in Fire Stance for the phys power boost just to increase the chance of landing it.
-You can't hit yourself with Whirlwind Teleport, so it should probably get nowarning=true in its targetting table. In vaults I tend to use in place as an area attack, this would remove the annoying confirmation prompt.
-Aura of Fire could probably stand to have a lower minimum self-damage, even going down .1 per talent point would be useful as 3% means I lose a bit over 100 HP when it triggers.

Edit: Forgot to mention that Elemental Destruction's description wasn't updated to match its new scaling, 10 points in it claims to add 100% resistance penetration when it actually gives 72%.

[Hirumakai]

Hi Stinkstink,

Thanks for the feedback.

-I'm not a fan of Purifying Fire having a max duration, considering it requires a Spell Save check and the enemies I most want to hit with it (stairbosses using Unstoppable or Swap Places/Evasion) tend to have 120+ spell save due to getting 22 levels in Spell Shield. I usually run around in Fire Stance for the phys power boost just to increase the chance of landing it.

This is an interesting observation. I've been doing a nightmare run through and near the end starting to notice a significant failure chance to apply against certain rand bosses. I had not really encountered it in normal games.

Apparently Wildfire/Cleansing flames just has its straight % chance (Which can get to 100% at 10/5 talent points). Disperse magic just works (no save it looks like, no check).

I could just break the symmetry with the other 3 Elemental strike talents and have it just apply, no save. This would put in line with Cleansing flames and disperse magic. Not sure why they don't, unless its because the removal effect is instant? In that case I could change it to instantly removing up to 3 things (which I guess is why disperse magic doesn't give a save, as well as wildfire - they have no duration to reduce), rather than doing it over 3 turns.

I'm hesitant to let the duration raise with no cap, as DarkGod had mentioned awhile ago when they scaled like stunning blow duration that it was a bit much on a silence effect, making it effectively better than the equivalent silence in the anti-magic toolkit, on an arcane user at that. And I happen to like the scaling symmetry between the strikes.

So I think a removal of up to 3 (instantly, no save) on a 8 turn cooldown, along with the other side benefits compared to Cleansing flames might be reasonable. Its probably better than letting it last longer than 8 turns (at which point refreshing it doesn't improve the removal speed), and as you say its nearly impossible to land anyways on certain bosses. Any opinions?

-You can't hit yourself with Whirlwind Teleport, so it should probably get nowarning=true in its targetting table. In vaults I tend to use in place as an area attack, this would remove the annoying confirmation prompt.

Good suggestion, will implement.

-Aura of Fire could probably stand to have a lower minimum self-damage, even going down .1 per talent point would be useful as 3% means I lose a bit over 100 HP when it triggers.

Sure, I can improve the high end scaling I think without affecting the campaign too much. And it really doesn't matter to NPCs. I can just put it on a scaling limit function down towards 0 at infinite points. I'll play around with it and see what looks good.

Edit: Forgot to mention that Elemental Destruction's description wasn't updated to match its new scaling, 10 points in it claims to add 100% resistance penetration when it actually gives 72%.

Good catch, will fix.

[stinkstink]

So I think a removal of up to 3 (instantly, no save) on a 8 turn cooldown, along with the other side benefits compared to Cleansing flames might be reasonable. Its probably better than letting it last longer than 8 turns (at which point refreshing it doesn't improve the removal speed), and as you say its nearly impossible to land anyways on certain bosses. Any opinions?

I actually like how Fire Strike isn't an instant sustain shutoff button the way Disperse Magic is, it allows the player opportunity to mitigate its effects or escape before a core ability is disabled. I'd prefer it to remain a temporary effect or create a Cleansing Flames-style field if possible.

[Hirumakai]

It would be possible to turn Fire strike into a ground debuff. The code for that basically already exists in the wildfire tree.

Here's a another option which isn't too far from the current version:

Since the talent duration doesn't scale after you get to 3 turns, I could replace that scaling with an application strength bonus. So basically as is now, but after you hit 3, get an additional 20% bonus per point.

So at an effective 5.0/5.0, no application bonus (as it is now).
If you're at an effective talent point of 10.0/5.0, it would double your current physical power to get the application strength (instead of having a 6 turn duration).
At 15.0/5.0 it would be triple physical power and so on.

This would then be applied to all 4 strikes. It kind of fits with the accuracy theme the class has going.

So in the case you present, I think your character has at least 160 physical power base from strength. Assuming you went with 10 raw points into fire strike, that 13 effective points, or 140% bonus. That would be another ~224 physical power for application purposes just from strength, which should compare well with the extra 198 raw spell save points 22 talent points in spell shield gets.

Thoughts?

[stinkstink]

That sounds like a clever solution that rewards investment beyond 5/5 but I'm not sure if your numbers were considering the phys power from Weapon Mastery. Before scaling that character has 271 physical power from strength and Weapon Mastery, 394 with Fire Stance up. Those numbers are 106 and 125 after scaling, and enemies with Spell Shield aren't too far ahead yet, tending to sit around 125-135 spell save at level 180-200 post-scaling. I have no clue how much wider that gap would be if I were playing a race with an experience penalty. It's also worth noting that I500 is a lot more generous with points post-50, so if I were playing vanilla I'd have a lot less freedom to throw everything into phys power as I've been doing.