Should it be 1- (.98^[12*shots per second]) since it's a 12 second proc? (It is 12 seconds right?)
I'm not completely convinced that, on average, the overlap will be in the middle of the previous proc...it should be more likely that you get an overlap soon after the previous proc, since getting an overlap late after the previous proc requires you to also not get an overlap soon after.
Maybe, we should try to calculate the time that Corrosive Salt is not procced? Essentially, calculate the chance, that in the the last 12 seconds, it didn't proc. (Or were you dong that...I'm not an expert at just understanding raw numbers ) That should be the percentage of time it's not procced. So, that chance is .98 raised to the power of the number of arrows that could be fired in those 12 seconds. Subtract it from 1 to find the % of time it is procced, Looks the same as what you've been doing so far...but now, just multiply that by a corrosive salt tic, and that will show your average damage per 2 seconds. I suppose you can divide that by the number of shots/swings in 2 seconds to see the average per shot/swing if you want. Hopefully that's right.
I guess it can get messed around with a bit with the matter of whether it starts ticking damage at the very start of the proc like Melf's Acid Arrow or after 2 seconds like Divine Punishment, if for example it's the latter, then an arbitrarily high attack speed (like, 12 monks all beating on a dragon with these handwraps) would give you almost no damage bonus since you would keep overlapping before the first tic can even do damage...and if it's the former, then it's possible to get more than one tic in a 2 second period...oh well. It's already messed up enough with bow attack speeds on a monk-cher constantly switching around...
I was mistaken, and it should be 70.3%.The wiki does state it is 12 seconds. I was (apparently?) thinking of the number of damages.Originally Posted by ZeebaNeighbaYou and me both! I'm not sure what I was thinking at the time, but I think did a center of mass kind of thing:I'm not completely convinced that, on average, the overlap will be in the middle of the previous proc...
1. It surely is equally likely to overlap at any given time in the 12 seconds, as dice have no memory.
2. So we have whatever% chance to overlap at 0.1 seconds, 0.2 seconds, 0.3 seconds, ... 11.7 seconds, 11.8 seconds, 11.9 seconds. (Or however far you wish to discretize it.)
3. Therefore the average will overlap duration/2 seconds.
I think past me has convinced myself, which stands to reason, but I understand if that's not convincing to non-mes.This is also a very important point, but again I think the practical concerns mitigate it already. We already feel that multiple Corrosive Salts are not good, so exactly how not good they are is not interesting because we won't be doing it anyway. Especially now that we are considering a proper duration of 12 instead of 6, a 2(ish) second delay doesn't seem too bad when we specify that only one person will use the Salt (in this case WoWo is Brad Pitt, naturally).I guess it can get messed around with a bit with the matter of whether it starts ticking damage at the very start of the proc like Melf's Acid Arrow or after 2 seconds like Divine Punishment, if for example it's the latter, then an arbitrarily high attack speed (like, 12 monks all beating on a dragon with these handwraps) would give you almost no damage bonus since you would keep overlapping before the first tic can even do damage...and if it's the former, then it's possible to get more than one tic in a 2 second period...oh well. It's already messed up enough with bow attack speeds on a monk-cher constantly switching around...
Basically, the way I'm saying it, sure dice have no memory. It has the same chance to proc in any specific second.
But there are less seconds where it's gone 11 seconds without overlapping than 1 second. Because for it to go 11 seconds without overlapping, it has to not overlap 11 times, which has a lower chance than simply not proccing for one second.
Think of my example in my last post, 12 monks with 23 BAB and Haste and zillion% doublestrike. Do you really think that 11 seconds can pass without it overlapping at all, and that it's just as likely as having it overlap one second after proccing?
The mathematically accurate way to calculate losses due to overlaps is as follows:
1) Determine attack speed in milliseconds and also the times in milliseconds after applying that each tick of C-Salt hits. For convenience I will skip this step and approximate the answers as 600ms attack speed (no double strike, sword and board with no glancing blows) and that C-Salt procs at 999, 2999, 4999, 6999, 8999 and 10999 ms.
2) Consider a specific C-Salt proc and determine the probability it reaches each tick of damage:
- the 999 ms tick: 98% (unless the attack at 600ms procs C-Salt, you will get this one)
- the 2999 ms tick: (0.98)^4 - To get to this tick, the 600, 1200, 1800 and 2400ms attacks all need to not proc a new C-Salt.
- 4999: (0.98)^7
- 6999: (0.98)^11
- 8999: (0.98)^14
- 10999: (0.98)^17
3) Convert the information in 2 into a probabalistic average damage per tick for a specific proc. i.e. if a C-Salt tick is 50+5d10 damage (average 77.5) then the 999ms tick is 'worth' 0.98*77.5 damage on average per proc, and the 10999ms tick is 'worth' (0.98^17)*77.5 damage.
4) Add the results in 3 for a total average damage per proc under the assumption of constant attacking.
I don't have a zerging problem.
I'm zerging. That's YOUR problem.
Wow - this thread blows me away. U guys r amazing !(genuine respect).
I realise this is probably not the correct thread but any chance any of u math guys can help a little dwarf twf WC bard out.
I have LitII Daxe, annihilation Daxe, Cacopony Daxe, Daxe with steam suffix and looking for Daxe with Wailing suffix.
Will be fatesinger ED with focus in sonic damage (harmonic resonance 10% per hit stacking sonic vulnerability up to 50% - 20 sec duration.
Twists of primal scream (sonic AoE), Anvil of Thunder (100 sonic/hit), and legendary tactics (aiming for stunning blow DC of 50).
Using light armour with Cacophonic guard ~9-10% of 100 sonic damage.
Guess what I am asking if anyone of u could give me a sense of DPS potential for current gear configurations - assuming STR 40, full TWF line.
Also any suggestions for alt gearing (ie twf with 2x Fang of Siberys) etc.
Big ask I know - np if no response. Math not my thing and just trying to see where stand DPS and if any obvious improvements can be made.
Tx
Guard stuff is dependent on mob attack speed which isn't known specifically afaik.
Most of what you need is available, in my sig is a link to a link regarding archer DPS, the link about attack speeds also has melee attack speeds, just work out your melee attack speed and multiple that by your average damage/hit and then account for crits and missing on a 1 and you're done.
Gid'day WoWo.
Luv it the way it sounds like making a model plane - part A then partB, combine all with part C and there she flies!
Tx for the feedback - some years since I did any math and there's a decent chance dwarf ale knocked off that part of my learning
Ill get on with it - Tx again.
i dont know if you noticed BUT. using a dummy to test ur damage is not a good idea. reason is. the first time you kill a dummy in a ship instance and the 2nd+ times are different. after the dummy has being destroyed once. and some one is still aboard the ship to prevent it from resetting. all other attacks on dummy deal an amount of 150%~ damage out put. check it out.
so the damage the O.P. put out from his corrosive. might have been over powered because of the added damage from "re-killing" the dummy. and not a normal damage output.
Last edited by masterzzan; 09-19-2012 at 05:26 AM.
^^^^^^^^
This Side UP
The Math, The math, My brain is in pain!!!!!
?_?
so it appears, the first thesis was proven to be not all exact.
I think #'s are just #, but when testing in game, that is more real, so i find whichever you personally think dishes out more damage is the one you would use.
2x disintegration wep's i think do more damage compared to that of a Lightning ---- weapon.
Reasons'
Disintegration vary(s)
Disintegration is untyped
and Lightning is pure electric
which is why i think that the first statement(thesis) is right
even though it is proven(calculated) wrong :/
[Summarizing up]
Post
This is a nice idea but it is very difficult to be objective about our own DPS. Even if asked to be objective about it I bet most people would still value their DPS as higher or lower depending on their mood, phases of the moon etc.
Good example of this is how many people think that high critical damage = good DPS. Big crits certainly make use feel good but they don't necessarily mean that we're doing more damage if the hits inbetween aren't as big. My opposite end of the scale character is my unarmed rogue who does huge amounts of damage/hit but has a rubbish critical profile.
Another point here is that people will think that they do a lot of DPS if they constantly pull aggro even if there are other factors involved like claw sets, hate stances or other forms of comparative threat generation (I'd basically say that wolf cloak is a required item for any high DPS character now).
So no, what you think does more damage is pretty irrelevant.
Nova, if you struggle with the resources I'm absolutely happy to help you along, maybe make a new thread about it to stop this one from getting too off topic?
you may be right about your post, but when is it that the math shown on the forums absolutely correct, because from what i see, game vs math is never correct because there is always going to be that value of error in a game.
Yes the math maybe correct, but accounting for errors is another case.
"saying no what you think does more damage is pretty... irrelevant... " is just your oponion, I'm sure others who do not read the forums could agree...
before the expansion even before the change to earth stance being into hate generation, i have rarely even fixed my eye's on hate generation, because the way i've seen it, was the first person to hit the monster has the aggro until he/she dies...
of course that observation has been seen in only the groups i have been in, i guess others may differ.
(but still, i always see the first person who attacks) has the aggro. and does not release the aggro...