Spiral Knights - Buff heavy swords

Buff heavy swords

14 Réponses [Dernière contribution]

dim, 04/12/2020 - 18:09

#6251

Scarlet-Green

Those are some excellent

Those are some excellent suggestion. I hope it's implemented.

mar, 04/14/2020 - 07:08

#6252

Sir-Pandabear

> The DA will deal less damage to neutral enemies than the hammer, while the Triglav/Sudarsuka will deal less than the DA

Suda and Hammer have identical damage, though. I fired up the game and checked a training hall to make sure. The only difference is, of course, that Rocket Hammer has the dash, which is more or less warranted between it being a DLC weapon, having pure specialized damage, and using an arguably riskier moveset.

On the wiki Suda states to have 288 damage at D28, but I'm fairly sure this is a typo as the earlier values are identical and the table hasn't been updated since I left. If you've measured this to be different that's fascinating and I'd love to know more about it.

Damage in Lockdown works the same as in PvE, with the exception that everyone has a minimum of 100 defense against all four types regardless of armour. For example, someone wearing skolver in D25 will have 140 normal defense and 125 piercing defense, but in LD they'll have 140 normal, 125 piercing, 100 elemental, and 100 shadow.

I agree that suda needs a buff. ~20% more damage overall and some sword dmg/ctr bonus to heavy armours would do the trick for me.

mar, 04/14/2020 - 11:31

#6253

Orangeo

Yeah double checking the

Yeah double checking the wiki, it's actually completely boneheaded. The DA's damage is stated to go down as you descend while the others go up. I actually recall it being different at last glance, but either way, less pure weapons should have better neutral scenarios. Them being better on elemental resistant enemies is not a genuine strength if they're only going to be used in neutral or advantageous situations.

Personally I think that, on top of the general damage changes suggested, the DA should have 10% more damage just to compete with the Rocket Hammer's purity. Even against just one elemental-weak enemy and a whole five neutral ones, assuming type-effective damage is twice as strong, this should give the DA the same average damage as the Rocket Hammer. That's 200+100*5 versus 150+110*5, which is a pretty fair way to weight it. Assuming the Suda with twice the normal damage sees twice the buff, the same comparison holds for one elemental weak enemy and four neutral ones.

I'm also not a big fan of how that LD damage works. Though to be honest I'm not even a fan of how normal works in PvE. It should just be treated as one-thirds each type.

Edit: as an inexpensive test, I went and compared a sealed sword to the 3* rocket hammer on slimes in both the garden of goo and red carpet runaround levels, and on both the T2 slimes and T3 ones that spawn for good measure. Neither sword did more damage than the other.

mar, 04/14/2020 - 11:11

#6254

Sir-Pandabear

For the first hit of DA, the damage numbers do go down in D28. This is because DA has two damage types and as such is countered by two defences.

However, that's only if you have no damage bonus. In practice it never comes up.

mar, 04/14/2020 - 11:15

#6255

Orangeo

Except that doesn't hold for

Except that doesn't hold for any other split damage weapon according to the wiki. Every brandish line gets increased damage as you descend.

mar, 04/14/2020 - 11:55

#6256

Sir-Pandabear

It's complicated! It only happens for a particularly area of damage number (see also: valiance/tortofist/strikers having 108 dmg through all of stratum 6)

What happens is that the damage increase per split doesn't keep pace with defense increase for monsters (for neutral damage). Any more damage (or with a damage bonus), then the damage keep up. Any less damage and the defense switches formulas to one that has a logarithmically growing impact.

It doesn't help that damage values do not grow at the same rate. I've been unable to find a specific formula for growth, so it appears that the curve for each damage value is manually tweaked. Despite or because of this, many weapons share damage numbers or the baseline of one. Some examples:

mar, 04/14/2020 - 13:59

#6257

Orangeo

Considering your "some

Considering your "some examples" is empty as I am now reading it, I'm inclined to think the wiki is just whack and that it's not actually that complicated. Nearly everything I've read on how defense works is prefaced with a "nope, nevermind!" followed by increasingly questionable test methodology.

To be honest I'd like a citation on how you've said defenses work in LD as well, because you're asserting that you get 100 free defense in whatever type you don't have. You're saying Sklover is 140 normal defense and 125 piercing? So, half Sklover and half Vog is 140 normal and 62.5 elemental and piercing? Which, in lockdown, would only be 140 normal and 100 of each other type, as opposed to the 140, 125, and 100/100 of pure Sklover? I would really hope Sklover isn't arbitrarily getting an extra 25 defense over a mixed setup. (Alternatively you meant only one piece of Sklover is 140/125, in which case a full set would be 280/250/100/100 versus a mixed 280/125/125/100, with the mixed set being a whole 125 points behind, an entire piece's defense bar.) Are you sure it's 100 in whatever you don't have, and not just 100 to everything?

mar, 04/14/2020 - 14:33

#6258

Sir-Pandabear

Oh, I wiped it as it was getting tangential. Forgot to remove the introduction. I found the thread where I first brought the weapon relations up:
https://forums.spiralknights.com/en/node/89924

> I would really hope Sklover isn't arbitrarily getting an extra 25 defense over a mixed setup.
It is. Don't buy shadow max skolver sets.

Unfortunately I can't cite anything, as I was the one who discovered it. I can link my own work, though:
Damage numbers for all weapons, all depths, including LD against various armour sets:
https://docs.google.com/spreadsheets/d/11myaO3o-cYxF-pX0ly6l_MRH54koq1Xb...

Lockdown calculator:
https://zeddidragon.github.io/sk-stats/

Sheet where I gathered damge in guild training hall:
https://docs.google.com/spreadsheets/d/1nC4xgGgZGAgufQwFH8w09Ik1S27wrHhq...
https://docs.google.com/spreadsheets/d/18AGiOG7NGGMn1E0RAjcWw3Ibxdeg8UoR...

Lockdown damage tests:
https://docs.google.com/spreadsheets/d/1EmACMXqCyWmvny4BuSbE1w8FSHwAC0pj...

I've tried with a load of sets, mixed or otherwise and have never observed less than 100 defense in T3. It's likely there so it isn't pointless to, say, deathmark a Black Kat before hitting them with Callahan.

mar, 04/14/2020 - 14:42

#6259

Sir-Pandabear

It's probably easier to exchange this information real time, so you can come with questions as soon as you have them and I don't just throw links all the time.

@ me (Zeddy) in the Arcade: https://discord.gg/8S8uMwg

mar, 04/14/2020 - 19:16

#6260

Orangeo

I mean a chart should just

I mean a chart should just speak for itself. Took me a second to figure out why you had "no armor" rows in the LD pages while also having numbers to the right of the attack names-themselves, but the former being 100 off from the latter made enough sense. Guns etc revealed that the green rows are deathmark (no defense) damage (which I would hope directly correlates with the tooltip information).

Do you have an explanation for why the DA's damage goes down as you descend a stratum for only neutral scenarios? And for why this is only the DA/GF and not weaker swords, which themselves should be less ahead on the defense curve? I'm starting to just think there is a typo not on the wiki but in the game.

mer, 04/15/2020 - 03:06

#6261

Sir-Pandabear

In short it's because the defense formula is complicated. I like to bring up Valiance because it's at the exact sweet spot where its damage doesn't grow at all. Meanwhile, guns doing less damage than Valiance doesn't shrink down the depths, and it's because at some point defense gives diminishing returns. This point is where defense > half the damage, at which points it switches from the simple formula of "damage - defense" to an obtuse logarithmic curve.

The reason it only happens for neutral is because neutral defense is twice that of weak defense, so the damage stays ahead of the curve when the curve is smaller. Resisted damage is already so far below the curve that the curve doesn't do much to it, if that makes sense. I'm not sure how to explain it better.

In the first chart I linked with damage values for all the depths you can see raw damage grayed out at the top of each row, along with enemy defense at the bottom of the table. You might find it enlightening to apply the defense formula (in the chart's function) at the depths and just see it in action.

Later edit - I dug up the formula:
damage * (1 - (1/2 + 0.19 * LOG10(((defence * 2) - damage)/15 + 1)));

On D24 enemies have about 65.5 base defense, and on D28 around 81.9. It interpolates linearly between those depths.

This means that a D28 zombie reduces elemental damage by ~82 and piercing/normal by ~163, but if the raw damage is below 326 (twice of the defence) then it's going to be reduced with the more complex formula shown above.

For split swords you chop the raw damage in half, then apply defense twice. Go ahead and crank the numbers for what happens to Divine Avenger on each depth, then you'll get the reducing numbers back out. It runs from 511 to 586 in Stratum 6.

Actually I'll just do it for you:
Depth 24

Raw: 511 / 2 = 255.5
Defence: 65.5 | 131
Weak:: 255.5 - 65.5 = 190
Neutral:: 255.5 * (1 - (1/2 + 0.19 + LOG10(( 131 * 2 - 255.5) / 15 + 1 )) = 120
Vs Undead:: 190 + 120 = 310
Vs Slime:: 120 + 120 = 240

In practice the real values are 1 damage off from these. This is because I we only know an integer approximation of the damage.

Depth 28

Raw: 586 / 2 = 293
Defence: 81.9 | 163.8
Weak:: 293 - 81 = 212
Neutral:: 293 * (1 - (1/2 + 0.19 + LOG10(( 163.8 * 2 - 293) / 15 + 1 )) = 117
Vs Undead:: 212 + 117 = 329
Vs Slime:: 117 + 117 = 234

Which is exactly what it says in the chart.

mer, 04/15/2020 - 06:25

#6262

Orangeo

>Resisted damage is already

>Resisted damage is already so far below the curve that the curve doesn't do much to it,
But it isn't just not doing much, it's going up. Is there a redundant attack increase on top of the defense changes?

Edit: Yes, so the base goes up in addition to defenses changing. This should just get removed.

mer, 04/15/2020 - 06:35

#6263

Zeddy

I think you're right about this one. Resisted values are probably backwards there and should be going downwards. I tried crunching some numbers and reproducing the case in the wiki and was unable to.

However, I can't answer it with certainty because I don't know how resisted damage works. There doesn't seem to be a factor that produces the expected results for different weapons.

mer, 04/29/2020 - 01:08

#6264

Xakepa

While a 10% dmg boost to

While a 10% dmg boost to troika line is cool, i'm not sure but I believe the swing speed is slower than sealed sword line, so at the end of the day SS will still outdamage troika even versus neutral enemies. What can be done is giving the troika line a unique bonus to differentiate it from SS, you could do that by e.g. giving slightly more range, which would fall in line with how the triglav/suda look longer than DA/GF. Maybe a 15% increase in range and 10% damage boost would make them more viable.

Recherche

Buff heavy swords