Corrosive Salt changes and calomel weapons comparison.

[Bufo_Alvarius]

Hmm seems like i might need to reconsider my alchie rapier. Was planning double water for velah but this kinda changes things a bit. Great info!

[LeLoric]

I'm getting the same for calomel, but you're lightning strike numbers seem off somewhere.

LitII hit:
9.5 base
7 holy
3.5 shock
---
20 *0.65 = 13

LitII crit:
19 base
12 seeker
7 holy
3.5 shock
11 burst/blast
---
52.5 *0.3 = 15.75

Shocking blast
14 * 0.05 = 0.7

Lightning strike:
610 @ 1.5% = 9.15 * 0.95 = 8.6925

Total:
13
15.75
0.7
8.6925
----
38.1425

Which is very close to the ~39 of the calomel. Even a small error on either crushing wave or corrosive salt's proc rates could push the LitII ahead.

Right found the error in litII calc. was multiplying the litII by .095 not .95 to account for the chance to miss on a 1. Extra zeroes suck.

This does put the litII back up there in consideration and completely wipes out the scimi versus khopesh comparisons LitII is generally better here regardless of x.

Thanks for pointing this out litII numbers seemed low but did this all while grading exams and playing the game today lol.

[Grailhawk]

litIIkhop := .75*(5.5+5+x+7+3.5)+.2*(3*(5.5+5+x+6)+7+3.5+11+11) +0.5e-1*14+610*(0.95e-1*0.15e-1);

or 33.719250 + 1.35 x

Maybe I'm miss understanding something but with the Lit2 Kopesh the math seams off. Using your set up I get

0.75*(5.5+5+x+7+3.5)+0.2*(3*(5.5+5+x+6)+7+3.5+11+1 1) +0.5e-1*14+610*(0.95e-1*0.15e-1);

0.75(21+x) + 0.20( 3(16.5+x) + 32.5)
15.75 + 0.75x + 0.20( 3x + 49.5 + 32.5) + 0.7 + 9.15
15.75 + 0.75x + 0.20(82 + 3x) + 0.7 + 9.15
15.75 + 0.75x + 0.60x + 16.4 + 0.7 + 9.15
15.75 + 1.35x + 16.4 + 0.7 + 9.15
32.85 + 9.15 + 1.35x
42 + 1.35x

What I'm confused about is what you'er doing with Lighting Strike why are you multiplying 0.095 by 0.015 it should simply be 610*0.015 which is the chance of a lightning strike given you hit.

[LeLoric]

So if somebody were making two alchemical weapons you could make a water water water and water air water or something of that nature. Would that be the highest dps or at least one weapon with the second tier air and the other with second tier water? What is your conclusions for alchemical weapon sets?

A lot depends on your mainhand dmg totals when doublestrike is concerned and also if theres any other source of doublestrike (epic spare hand etc.) That leads us to comparing entire gear sets though. Also remember dual corrosive salts do up the chance of losing avg dps through reapplications of corrosive salt/crushing wave before a previously applied one has expired.

Alchem combos are really hard to say one is better than the other which is good as there is a fine balance amongst them then.

For tier 1 fire does the most pure dps but is also highly resisted/immune. The other three provide additional pseudo dps through helplessness inducing states to that is hazrd to quantify. I personally like the earth ones, they provide both a reflex and a fort based helplessness inducer. My endgame play is varied enough right now that on specific element doesn't seem to be resisted more than any other.

For a pair of all purpose dps alchems I think I would probably go earth/water/whatever in mainhand and air/air/whatever offhand for anything with stunning blow otherwise offhand would be earth/air/whatever. I say whatever because tier three is essentially the same dps but with different dmg type. Choose your dmg type of choice here.

[LeLoric]

Maybe I'm miss understanding something but with the Lit2 Kopesh the math seams off. Using your set up I get

0.75*(5.5+5+x+7+3.5)+0.2*(3*(5.5+5+x+6)+7+3.5+11+1 1) +0.5e-1*14+610*(0.95e-1*0.15e-1);

0.75(21+x) + 0.20( 3(16.5+x) + 32.5)
15.75 + 0.75x + 0.20( 3x + 49.5 + 32.5) + 0.7 + 9.15
15.75 + 0.75x + 0.20(82 + 3x) + 0.7 + 9.15
15.75 + 0.75x + 0.60x + 16.4 + 0.7 + 9.15
15.75 + 1.35x + 16.4 + 0.7 + 9.15
32.85 + 9.15 + 1.35x
42 + 1.35x

What I'm confused about is what you'er doing with Lighting Strike why are you multiplying 0.095 by 0.015 it should simply be 610*0.015 which is the chance of a lightning strike given you hit.

I have already fixed that. 610*.015 is your dmg per hit not dmg per swing. The multiplier to change it to dmg per swing was supposed to be .95 not .095 which is where the error was.

[Kinerd]

Nice! Two things:

My first tests were done with the info listed on the wiki for corrosive salt at 4 tics of 20d6
(70 avg) was testing something else on my rogues epic smallblade and noticed that it was providing 6 ticks where it used to give 4. Couple that with information that I received about the proc rate 2% (same as wiki) and dmg being 5d10+50. Testing confirms this dmg value as more likely than the 20d6 as the range of numbers received is too tight to be 20d6.

I have seen similar information, I just wanted to point out that 20d6 and 5d10 have very similar standard deviations: 7.64 for the first and 6.42 for the second. All the observed values fall within two standard deviations of both distributions, so the data alone is not enough to confirm either way. (Not that it matters all that much.)

Second, you mention proc overlaps, and at 6 seconds it's certainly getting into the territory where overlaps are plausibly significant. Have you observed what specifically happens? Does the duration refresh or do you just not get processes?

[LeLoric]

Nice! Two things:I have seen similar information, I just wanted to point out that 20d6 and 5d10 have very similar standard deviations: 7.64 for the first and 6.42 for the second. All the observed values fall within two standard deviations of both distributions, so the data alone is not enough to confirm either way. (Not that it matters all that much.)

Second, you mention proc overlaps, and at 6 seconds it's certainly getting into the territory where overlaps are plausibly significant. Have you observed what specifically happens? Does the duration refresh or do you just not get processes?

Durations refresh on proc overlaps. Yes the overlaps can be significant. The later in the duration of the dot the less dmg lost too.

As for the dice numbers yes the SD is very close but the ranges of the two rolls are much different. 5d10 plus 50 has a range of 55-100 whereas 20d6 has a range of 20-120. That being said the outliers for a corrosive salt under a 20d6 would be much further from the mean. Observation has shown that there are no such outliers. now a weighted 20d6 like casters get actually being 20d3+60 would be a possibility but then we get a mean which does not match observed numbers.

[Kinerd]

Durations refresh on proc overlaps. Yes the overlaps can be significant. The later in the duration of the dot the less dmg lost too.

Interesting, interesting, let's see if I can whip a little math at this. I'll use someone vanilla TWFing these weapons first. At 1.8 proc chances per hit, 19 hits per 20 swings, 102.2 swings per minute, we have 2.9127 proc chances per second. After 6 seconds we will have seen at least one proc:
1 - .98 ^ (6 * 2.9127)
= 29.7% of the time.

Hence, 29.7% of procs during continuous combat will overlap with a previous proc (outside of the initial 6 seconds of combat). If we estimate that on average a refreshed proc will only contribute 3 seconds, then our practical DPS contribution from a Corrosive Salt will be:
.02*(5*5.5+50) * (6 * 71.3% + 3 * 29.7%)
= 8.01

Rather than the 9.3 we would get if they stacked. Using only one Corrosive Salt weapon (in the main hand) works out to 8.47 practical DPS. Rather than gaining 0.8 * 9.3 = 7.44 from the off hand Corrosive Salt, we only gain 1.8 * 8.01 - 8.47 = 5.95, so if the off hand Saltless weapon is less than 1.49 nominal DPS behind the Salt weapon, you're better off Saltless. To illustrate, if a weapon had Lightning Strike or Corrosive Salt but was otherwise identical, you would want a Corrosive Salt in the main hand but a Lightning Strike in the offhand, because:
0.8 * 9.3 - 1.49 < 0.8 * 9.15 * 19/20

Faster rate of attack, whether through animation speed or additional processes, further dilutes the effectiveness of a second Corrosive Salt.

[Monkey_Archer]

Interesting, interesting, let's see if I can whip a little math at this. I'll use someone vanilla TWFing these weapons first. At 1.8 proc chances per hit, 19 hits per 20 swings, 102.2 swings per minute, we have 2.9127 proc chances per second. After 6 seconds we will have seen at least one proc:
1 - .98 ^ (6 * 2.9127)
= 29.7% of the time.

Hence, 29.7% of procs during continuous combat will overlap with a previous proc (outside of the initial 6 seconds of combat). If we estimate that on average a refreshed proc will only contribute 3 seconds, then our practical DPS contribution from a Corrosive Salt will be:
.02*(5*5.5+50) * (6 * 71.3% + 3 * 29.7%)
= 8.01

Rather than the 9.3 we would get if they stacked. Using only one Corrosive Salt weapon (in the main hand) works out to 8.47 practical DPS. Rather than gaining 0.8 * 9.3 = 7.44 from the off hand Corrosive Salt, we only gain 1.8 * 8.01 - 8.47 = 5.95, so if the off hand Saltless weapon is less than 1.49 nominal DPS behind the Salt weapon, you're better off Saltless. To illustrate, if a weapon had Lightning Strike or Corrosive Salt but was otherwise identical, you would want a Corrosive Salt in the main hand but a Lightning Strike in the offhand, because:
0.8 * 9.3 - 1.49 < 0.8 * 9.15 * 19/20

Faster rate of attack, whether through animation speed or additional processes, further dilutes the effectiveness of a second Corrosive Salt.

Hmm... assuming both weapons are equal dps, to get the most out of a single corrosive salt weapon, we'd want it in our offhand. 20% less hits = 20% less procs which means less chance to overlap (roughly 19% chance for a single mainhand to overlap, and 15% for a single offhand to overlap) Whereas a direct proc weapon like Lightning strike gets 100% effectiveness in either hand. Given that the calomel is roughly equal dps to other standards (HOBGB, Lit2, etc..) the best setup seems to be calomel offhand + HOBGB mainhand (lit2 is questionable)

The question I have now is whether a calomel rapier or an epic smallblade is a better offhand... but I need coffee before that...

[Monkey_Archer]

Ok, so the smallblade vs calomel offhand is relatively simple. Both have the same base damage and crit profile, both have corrosive salt, and both have a D4 damage effect...

So the only difference is:

Calomel
1.35 tidal burst effect (4.5*6/20)
5.168 crushing wave (5.44*19/20)
-----
6.518 * 0.8 offhand = 5.2144

Smallblade
4.275 bleed * 0.8 offhand = 3.42

So the smallblade passes the calomel due to doublestrike when your average mainhand damage is greater then ~30 = (5.2144-3.42) / 0.06, and would pass the calomel vs dragons when mainhand > ~275 (but thats not happening)

[LeLoric]

Bad news.

In doing some testing with a guildie not only do your own proc's reset the timer another person using a corrosive salt weapon if theirs procs will reset the timer and you lose the rest of your ticks. Same with the crushing wave, slicing winds, magma surge.

This can lead to a lot of overlapped lost dps in groups and especially raids if lots are running dot weapons.

The sheer number of corrosive salt weapons now out there with greensteel, smallblades, calomel, and alchem tier 2 water means it is more likely to have multiples of these in a group.

If everyone in your edragon group is running calomels theres a lot of lost dmg potential there.

Corrosive salt is still awesome but just be wary that if too many people are running them you can lose a lot of dmg output.

This puts a lot of concern into running dual salt weapons too. It may be better to pair a salt weapon with a non salt weapon to reduce this chance of lost dps.

[Monkey_Archer]

Bad news.

In doing some testing with a guildie not only do your own proc's reset the timer another person using a corrosive salt weapon if theirs procs will reset the timer and you lose the rest of your ticks. Same with the crushing wave, slicing winds, magma surge.

This can lead to a lot of overlapped lost dps in groups and especially raids if lots are running dot weapons.

The sheer number of corrosive salt weapons now out there with greensteel, smallblades, calomel, and alchem tier 2 water means it is more likely to have multiples of these in a group.

If everyone in your edragon group is running calomels theres a lot of lost dmg potential there.

Corrosive salt is still awesome but just be wary that if too many people are running them you can lose a lot of dmg output.

This puts a lot of concern into running dual salt weapons too. It may be better to pair a salt weapon with a non salt weapon to reduce this chance of lost dps.

Interesting...
Guess i'm just going to stick with smallblade offhands for all my pierce spec characters....

[LeLoric]

Interesting...
Guess i'm just going to stick with smallblade offhands for all my pierce spec characters....

Yeah actually my plan until i get more tier two alchems then id probably go with a tier two water and tier two air combo.

I'll still make up some calomels too They are exceptional ele beaters and good in dragons.

[Kinerd]

Bad news.

In doing some testing with a guildie not only do your own proc's reset the timer another person using a corrosive salt weapon if theirs procs will reset the timer and you lose the rest of your ticks. Same with the crushing wave, slicing winds, magma surge.

This can lead to a lot of overlapped lost dps in groups and especially raids if lots are running dot weapons.

The sheer number of corrosive salt weapons now out there with greensteel, smallblades, calomel, and alchem tier 2 water means it is more likely to have multiples of these in a group.

If everyone in your edragon group is running calomels theres a lot of lost dmg potential there.

Corrosive salt is still awesome but just be wary that if too many people are running them you can lose a lot of dmg output.

This puts a lot of concern into running dual salt weapons too. It may be better to pair a salt weapon with a non salt weapon to reduce this chance of lost dps.

Bad news indeed. Oh well, thanks for testing and reporting.

[dTarkanan]

How do the level 16 Calomels with a Regular Dragon Bane and a 1d6 elemental compare vs Lit 2s? How about with a 2d6 vicious?

[destiny4405]

i'm not sure from you're formulaes, but did you assume damage from lifedrinker effect (25-45 damage on nat 20)?

[LeLoric]

i'm not sure from you're formulaes, but did you assume damage from lifedrinker effect (25-45 damage on nat 20)?

0.5e-1*35 yep that's it right there.

[Warinx]

Nice work. +1
Would be interesting to know how the Calmoel Falchion compares to the other more obtainable wepaons like the Fire Axe and the Antique Greataxe or LitII Falchion

[wax_on_wax_off]

Interesting, interesting, let's see if I can whip a little math at this. I'll use someone vanilla TWFing these weapons first. At 1.8 proc chances per hit, 19 hits per 20 swings, 102.2 swings per minute, we have 2.9127 proc chances per second. After 6 seconds we will have seen at least one proc:
1 - .98 ^ (6 * 2.9127)
= 29.7% of the time.

Hence, 29.7% of procs during continuous combat will overlap with a previous proc (outside of the initial 6 seconds of combat). If we estimate that on average a refreshed proc will only contribute 3 seconds, then our practical DPS contribution from a Corrosive Salt will be:
.02*(5*5.5+50) * (6 * 71.3% + 3 * 29.7%)
= 8.01

Rather than the 9.3 we would get if they stacked. Using only one Corrosive Salt weapon (in the main hand) works out to 8.47 practical DPS. Rather than gaining 0.8 * 9.3 = 7.44 from the off hand Corrosive Salt, we only gain 1.8 * 8.01 - 8.47 = 5.95, so if the off hand Saltless weapon is less than 1.49 nominal DPS behind the Salt weapon, you're better off Saltless. To illustrate, if a weapon had Lightning Strike or Corrosive Salt but was otherwise identical, you would want a Corrosive Salt in the main hand but a Lightning Strike in the offhand, because:
0.8 * 9.3 - 1.49 < 0.8 * 9.15 * 19/20

Faster rate of attack, whether through animation speed or additional processes, further dilutes the effectiveness of a second Corrosive Salt.

How do I work this out for a bow especially considering 3 different attack speeds?

Edit: wiki says 5% proc rate for crushing wave, is that incorrect?

[Kinerd]

I always freak out a little when someone quotes me from a time I can't distinctly remember. Who was that person that said what I said? Argo! (Not that Argo.) Identity! Locke cannot save me, let's see.

.

The first thing is apparently to calculate out your proc chances per second. That seems pretty reasonable.

Then I did 1 - .98 [the chance of not processing] ^ ( 6 [duration in seconds] * # [proc chances per second] ), which gives us the % of procs that will overlap with another.

Then for reasons I do not currently understand I estimated that an overlapped proc would last 3 seconds, or half the nominal duration.

.

So for a bow, the only thing that changes is the first step. I assume by 3 different attack speeds you are referring to manyshot, 10k shot, and plain ol' shot. When the durations of these are 20, 30, and 10(?) seconds respectively, I think this makes it exceedingly complicated to calculate a 6 second duration effect directly. Rather than do that, I would advise you to obtain an average attack speed: 1/6 manyshot, 1/2 10k, 1/3 plain (or normalize to 75% instead of 100% if you're regenning ki with unarmed). Regardless of the specifics, getting to a single attack speed is the key. It underestimates the loss of stacking in the fast period and overestimates it in the slow period, but I sincerely doubt it will net do so to a significant level.

.

As to the wiki, I was dealing with Corrosive Salt rather than Crushing Wave. As it happens, though, I hear that Crushing Wave is "2% for 4d6+20 cold damage and 4d6+20 bludgeon damage every 2 seconds for 8 seconds".