Sensuki's Suggestions #007: Might Percentile bonus in relation to Weapon Balance

[Sensuki]

You guys are forgetting recovery time when talking about weapon speed.

 

A Stiletto does 12-20 damage, and makes an attack every 50 frames, at 30 frames a second (animation speed is 30FPS) so it attacks every 1.666667 seconds

 

A War Hammer does 16-26 damage, and makes an attack every 60 frames, so it attacks every 2 seconds.

 

This is of course, assuming you are not wearing armor or a shield.

 

Two Weapon Fighting I believe, reduces the overall damage you do with both weapons, but I am not sure about the exact mechanics.

 

The math works out so that the 1H Normal weapon has a higher flat DPS, but the special properties of the 1H Fast weapons should make them an interesting choice if they are done right.

 

A small weapon benefits *more* from strength than a large one, because it suffers more from DT.

I don't think that's actually true.

Edited August 21, 2014 by Sensuki

[Caerdon]

 

A small weapon benefits *more* from strength than a large one, because it suffers more from DT.

I don't think that's actually true.

 

Weapon X deals 20 damage per hit. Against armor DT of 10 it only deals 10 damage. With 50% might bonus it deals 20 damage against that armor. It's effective DPS has increased 100%.

 

Weapon Y deals 40 damage per hit. Against the same armor it deals 30 damage. With 50% might bonus it now deals 50 damage. It's effective DPS has increased 67%.

 

It's as simple as that.

[Sensuki]

It's not as simple as that man, it's a lot more complicated. Do the math with the actual weapon damage ranges - on grazes, on hits and on crits and take into account the actual DPS (I provided you the times in a post above) - and tell me that.

 

You also can't get a 50% Might bonus. Calculate it with realistic Might values such as 10 (20%), 15 (30%) and 20 (40%).

Edited August 21, 2014 by Sensuki

[Answermancer]

Sensuki he wasn't saying that the weapon damage is fine (I suspect it's true that the weapons need a balance pass and 2-H weapons do too much damage).

 

He was just saying that weaker weapons benefit more (in terms of % damage gained) from Might vs. DT than large ones because Might helps them overcome DT that would otherwise eat up a significant amount of their damage.

 

With a more "realistic" example, 30% bonus from Might, 10 DT:
20 damage base (-10DT = 10 Damage Dealt, no Might bonus)
20 * 1.3Might = 26 - 10DT= 16 Damage
16 vs. 10 is a 60% increase in damage dealt.

 

40 damage base (-10DR = 30 Damage Dealt, no Might bonus)
40 * 1.3Might = 52 - 10DT = 42 Damage
42 vs. 30 is a 40% increase in damage dealt.
 

[Caerdon]

It's not as simple as that man, it's a lot more complicated. Do the math with the actual weapon damage ranges - on grazes, on hits and on crits and take into account the actual DPS (I provided you the times in a post above) - and tell me that.

 

You also can't get a 50% Might bonus. Calculate it with realistic Might values such as 10 (20%), 15 (30%) and 20 (40%).

 

I was using simplified numbers to get my point clearly across. I assure you, changing them to more realistic ones won't change the result, and neither will accounting for grazes and crits. As long as we have a reasonable DT (i.e. not so high that only the large weapon with full might bonus can strike through at all), the weapon dealing less damage per hit benefits relatively more from higher might.

 

Attack rates don't matter one bit if we're talking about relative increases of effective DPS, which is usually the most sensible way as it's the single most important factor in weapon balancing.

[Sensuki]

Okay sorry, I think I've been trying to explain this from the wrong angle.

 

You are correct that the percentile benefit increase from Might is better on the lower weapons, a 60% increase in per hit damage is higher than a 40% increase in per hit damage. However due to the fact that the weapon has less damage anyway, the fact that you are getting 60% from the lower weapon and only 40% from the higher weapon is cancelled out by the actual amounts of damage that they do, regardless of the percentage increase.

 

You are getting more flat real damage from the weapons that do higher damage, even though the 60% is higher than the 40%, taking into account the actual weapon damage, the benefit (IMO) of Might on lower damage and higher damage weapons is actually the same.

 

The 60% increase in your example and the 40% increase example give the same amount of DPS increase in relation to the weapon damage (30% increase)

 

Does that make any sense?

Edited August 21, 2014 by Sensuki

[Falendor]

My 2% cents.

a % increase to DPS is a % increase to DPS.

hypothetical

Dagger: attack 2 per sec for 10 dmg = 20 dps

sword: attack 1 per sec for 20 dmg = 20 dps

With 30% might added in

dagger: attack 2 per sec for 13 = 26 dps

sword: attack 1 per sec for 26 = 26 dps

 

its the in game weapons that are off balance right now not the fundamental math the stats work off of.

 

P.S. might is awesome with high base damage weapons if you are trying to set up some sort of "alpha strike" strategy.

[Matt516]

While your suggestions are usually quite stellar, Sensuki, I think you're off base here. Percentage damage increase is actually the only way to make damage bonuses not favor one weapon style or the other. Flat damage increases will seriously favor any weapon that strikes more.

[Matt516]

Okay sorry, I think I've been trying to explain this from the wrong angle.

 

You are correct that the percentile benefit increase from Might is better on the lower weapons, a 60% increase in per hit damage is higher than a 40% increase in per hit damage. However due to the fact that the weapon has less damage anyway, the fact that you are getting 60% from the lower weapon and only 40% from the higher weapon is cancelled out by the actual amounts of damage that they do, regardless of the percentage increase.

 

You are getting more flat real damage from the weapons that do higher damage, even though the 60% is higher than the 40%, taking into account the actual weapon damage, the benefit (IMO) of Might on lower damage and higher damage weapons is actually the same.

 

The 60% increase in your example and the 40% increase example give the same amount of DPS increase in relation to the weapon damage (30% increase)

 

Does that make any sense?

 

You're right, a lower damage weapon would get less flat damage from the percentage increase. But why were you using that lower damage weapon in the first place? Presumably, it attacked faster. If we take two weapons: one low damage with high attack speed, and one high damage with low attack speed - but both having the same DPS? Then a percentage increase will affect the DPS of both weapons in the same way.

 

All other things equal, the percentage increase does not favor any weapon types when you look at it from a DPS perspective (which is the only sensible way to look at it).

 

Now, with armor and damage reduction.... there may be an argument to be made that the higher damage weapon benefits more because less attacks total = less chances for the armor to reduce damage. But I'm not 100% clear on how armor is implemented in PoE so I don't want to make big statements about it.

[Answermancer]

Okay sorry, I think I've been trying to explain this from the wrong angle.

 

You are correct that the percentile benefit increase from Might is better on the lower weapons, a 60% increase in per hit damage is higher than a 40% increase in per hit damage. However due to the fact that the weapon has less damage anyway, the fact that you are getting 60% from the lower weapon and only 40% from the higher weapon is cancelled out by the actual amounts of damage that they do, regardless of the percentage increase.

 

You are getting more flat real damage from the weapons that do higher damage, even though the 60% is higher than the 40%, taking into account the actual weapon damage, the benefit (IMO) of Might on lower damage and higher damage weapons is actually the same.

 

The 60% increase in your example and the 40% increase example give the same amount of DPS increase in relation to the weapon damage (30% increase)

 

Does that make any sense?

 

Yeah, I get what you're saying, it's certainly a good point and something for them to balance around, but like others have said attack speed SHOULD equalize this a bit.

 

Perhaps what we need is for small weapons to be much, much faster, if you got 2-3 times as many attacks with a small weapon as you do with a big one, then they'd be pretty appealing I think, especially since the percentile benefit of might on those weapons is better.

 

Using my hypothetical math from above (final values of 16 Damage and 42 Damage after Might bonus and DT subtraction), a light weapon would be as good as the heavy one if it attacked 2.625 times for every one heavy weapon attack. That's a big disparity but:

A. Fake numbers

B. I kinda think that attacking with a couple of daggers should be much faster than with a greatsword

Edited August 22, 2014 by Answermancer

[Sensuki]

While your suggestions are usually quite stellar, Sensuki, I think you're off base here. Percentage damage increase is actually the only way to make damage bonuses not favor one weapon style or the other. Flat damage increases will seriously favor any weapon that strikes more.

 

Quite correct, this is the worst one I've made by far and I already changed my mind half way through the thread after doing some proper calculations.

 

You're right, a lower damage weapon would get less flat damage from the percentage increase. But why were you using that lower damage weapon in the first place? Presumably, it attacked faster. If we take two weapons: one low damage with high attack speed, and one high damage with low attack speed - but both having the same DPS? Then a percentage increase will affect the DPS of both weapons in the same way.

 

All other things equal, the percentage increase does not favor any weapon types when you look at it from a DPS perspective (which is the only sensible way to look at it).

 

Now, with armor and damage reduction.... there may be an argument to be made that the higher damage weapon benefits more because less attacks total = less chances for the armor to reduce damage. But I'm not 100% clear on how armor is implemented in PoE so I don't want to make big statements about it.

 

I'm not 100% sure whether I have the recovery times right. I've been calculating with a flat 1 second recovery unarmored for everybody. The recovery period may actually just be the same as the weapon speed, I am not sure.

Edited August 22, 2014 by Sensuki

[Karkarov]

Just to throw this out there but I bothered looking into it yesterday and yes DT = a straight number.  10 DT means you do your damage, then the game minuses 10, and that is the real damage you ultimately did.  That said I think there are situations where DT either doesn't apply or can be heavily bypassed.  In practice I have run into DT being a problem far less than damage type resistance.  For example I made a post about how OP a Cipher with high might and the Justice sword is.  I was literally landing 120 damage crits to start every fight.  But then I ran into the Crystal Spiders who are highly resistant to slashing damage and suddenly I was only hitting for like 5-10 damage.  Meanwhile BB Priest with a one hand war hammer was doing far more damage despite the DT because there was no crushing resistance on the spider.

[Sensuki]

Doesn't the Justice Sword do like 1000% Crush Damage? or is that just a tooltip error.

 

That would be funny if you were a Druid and selected Crush as your Wildpaws or whatever it's called. Whooping damage!

Edited August 22, 2014 by Sensuki

[Matt516]

The DT's magnify the effect too.

 

I'm not entirely sure the combat math holds up to tell you the truth. Anything with a DT makes light, fast weapons as good as useless; they're reduced to decimal-point grazes basically, making slow, heavy weapons not just, say, 50% or even twice as effective, but, like, 10, 20 times as effective. Whereas slow, heavy weapons are still reasonably good against enemies with low DT, where those fast light ones would in fact be better.

 

That makes light, fast weapons highly unattractive. It's a choice between something that's really good against a small subset of critters and useless against the rest, and something that's really good against the hardest-to-damage critters and OK against the rest. Bit of a no-brainer there, especially if you only have two weapon slots so there's no room for, say, light-melee, light-ranged, heavy-melee, and heavy-ranged sets.

 

Come to think of it, four slots would mitigate the problem. I've nothing against playing RPS with weapons if that's how Josh rolls, but in that case I would like to have enough slots so that I wouldn't have to fiddle with the inventory all the time.

 

I do notice that I've already switched to the heaviest weapons I can find and everything is suddenly a lot easier. I'm sure there are some situations where it'd be better to switch to something else but nothing that's giving me real trouble.

 

So unless I'm missing something major, weapon balance is off. Maybe tune the DT's so that light weapons aren't so completely useless.

 

I believe the advantage of fast, light weapons would be more chances to score interrupts, as well as (hopefully) DT penetration on the really pointy ones. You may still be right re: whether the math holds up, but I don't think it's quite that cut and dry. In a well balanced system, maybe fast weapons would have more raw dps (better against unarmored) and slow weapons more single hit damage (better against armored)? We also have to take the damage types into account - DT is only effective for most armors against certain types of damage. So with a fast mace and slow sword, both having equal dps, striking against plate, the mace should do more damage.

 

It's all tuneable, really. You're completely right that just taking DT into account slow heavy weapons seem to win out. But I think there are more factors in play that (if tuned properly by people better at this than I am) can make the system balanced overall - factors like damage type, raw dps differences, skills, etc.

[Karkarov]

Doesn't the Justice Sword do like 1000% Crush Damage? or is that just a tooltip error.

 

That would be funny if you were a Druid and selected Crush as your Wildpaws or whatever it's called. Whooping damage!

It does, but it is a slashing weapon so doesn't actually do crushing damage ;p.  It is either a tooltip error or some sort of inside joke.

[Sensuki]

Yeah, which is why if it was actually mechanically built in that would be hilarious as a Druid with crush as their bonus damage.