Curse

Chthon · Jul 13, 2009, 04:23 AM // 04:23

Quote:

Originally Posted by traversc

snip

This is dead wrong. I'm going to try one more time, then just declare you a hopeless idiot and give up. Since MoP's complexities are obviously confusing you, I'm going to try to build up from simpler examples that highlight the basic math of skill DPS.

Let's start with a much more simple spell: Flare. Flare's DPS equals damage-per-cast divided by recycle time. Any problem there? No? Good.

Now let's move on to a slightly more complicated example: Fireball. Fireball's DPS equals damage-per-cast times the average-numbers-of-foes-hit divided by recycle time. Any problem there? No? Good.

Now let's take an example that highlights your error: IAtS. IAtS's DPS equals damage-per-trigger times number-of-triggers divided by recycle time. Any problem there? No? Good.

Now, does increasing your hit rate increase IAtS's DPS? No, it doesn't. Unless your hit rate is so slow that IAtS recharges before your run out of triggers, you're going to get the same total damage over 1 recycle period regardless of how fast you burn through your triggers. Got that?

At a very low granularity, burning through all your triggers in short order can look like a DPS increase ("OMG Barrage just did +140 in 1 sec!!"), but it's really just a spike. Your overall DPS remains constant once you look at a period of time at least as long as a recycle period. Not that spike's aren't nice, but they aren't DPS.

Now, how does this relate to your mistake about MoP? You fail to see that MoP, like IAtS, has a limited number of triggers per cycle. So, just like IAtS, burning through your triggers quickly or slowly has no effect on DPS over the long haul. Also like IAtS, the only exception is going to be when your hit rate is so low that you're not getting all the triggers you could be (in MoP's case that means failing to kill the target before MoP expires).

It may be easier for you to conceptualize this by temporarily ignoring some of MoP's unrelated complexities. Imagine a simplified version of MoP named proto-MoP that does X damage to one adjacent foe each time the target takes any kind of damage. What is proto-MoP's DPS? It equals damage-per-trigger times number-of-triggers divided by recycle time -- the same as IAtS. But now we're in a position to expand the number-of-triggers variable into something more useful. Number-of-triggers equals monster-life divided by average-damage-packet-size. So we can substitute that into our first equation and come out with: proto-MoP's DPS equals damage-per-trigger times monster-life divided by average-damage-packet-size divided by recycle time.

Stop right there. Reread those last two paragraphs. This is what you need to understand to understand your error.

Sit down and work through a simple mental experiment: Imagine two adjacent practice dummies. The time it takes you to autoattack a dummy to death equals one recycle period for proto-MoP. Your damage per auto-attack equals twice proto-MoP's damage per trigger. Imagine casting proto-MoP on one dummy, auto-attacking it to death, thereby taking off half the other dummy's life. Now work out the DPS from proto-MoP over 1 cycle period. Repeat the experiment with a friend also auto-attacking the first dummy. Now work out the DPS from proto-MoP over 1 cycle period. Is it any different? It shouldn't be. Repeat this exercise slowly in your head (or on paper) until you get it correct.

Once you've gotten your mind wrapped around the fact that MoP has a limited number of triggers per cycle and how that plays out in the DPS equation, putting the irrelevant complexities back in place isn't so hard: Unlike proto-MoP, the real MoP has restrictions on what triggers it. The number of triggers you get is, quantity, monster-life minus life-lost-to-non-physical-damage, close quantity, divided by average-physical-damage-packet-size. (There's the anti-synergy with Discord.) Unlike proto-MoP, the real MoP hits all adjacent foes, so the damage-per-trigger needs to be multiplied by average-number-of-foes-hit.

Put them altogether and you get damage-per-trigger times average-number-of-foes-hit times, quantity, monster-life minus life-lost-to-non-physical-damage, close quantity, divided by average-physical-damage-packet-size divided by recycle time -- which is the formula from my last post.

Now, I'm done posting on this topic. You can put in the mental work to understand your error before you post again, or you can choose not to -- I really don't care. If you choose not to, I'm not going to waste any more time trying to correct you.

Jeydra · Jul 13, 2009, 08:33 AM // 08:33

Quote:

Originally Posted by traversc

.... There obviously SHOULD be.

He's arguing that 1 minion triggering MoP 100 times is just as good as 100 minions triggering MoP 1 time each.

Seriously?Seriously?Seriously?Seriously?Seriously? Seriously?Seriously?Seriously?

He didn't say it's just as good. He said they trigger MoP the same number of times. Read it again.

Btw Chton I didn't read all you and traversc wrote, but this in particular is certainly wrong.

Quote:

Originally Posted by Chton

All put together, the relationship looks something like this:
Average DPS = (42*A*((B-C)/D))/E
Where:
A = average number of adjacent foes
B = average hp of target when MoP is cast
C = average total damage done to target via non-physical damage
D = average packet size of physical damage done to target
E = average effective cast+recharge time for MoP considering effect of AP

Number of minions isn't a variable there.

I won't go into the intrinsics. I'll just examine the result. Of course the number of minions is a variable! It's not directly encoded in your equation, but it definitely is a factor. The reason is so obvious I'm surprised you haven't already seen it.

The more minions you have, the faster monsters die. The faster monsters die, the more times you can trigger AP in unit time. The more times you trigger AP in unit time, the lower E is. The lower E is, the higher the DPS is!

Q.E.D.

Like I said, this is so obvious I'm impressed you haven't seen it, and that's not considering quickly triggering MoP means mobs have less time to scatter ...

AtomicMew · Jul 13, 2009, 05:30 PM // 17:30

Quote:

Originally Posted by Jeydra

He didn't say it's just as good. He said they trigger MoP the same number of times.

That is patently false. MoP does NOT trigger the same amount of time. Refer to my previous post and below.

Quote:

Now let's take an example that highlights your error: IAtS. IAtS's DPS equals damage-per-trigger times number-of-triggers divided by recycle time. Any problem there? No? Good.

Now, does increasing your hit rate increase IAtS's DPS? No, it doesn't. Unless your hit rate is so slow that IAtS recharges before your run out of triggers, you're going to get the same total damage over 1 recycle period regardless of how fast you burn through your triggers.

IATS is not analogous to MoP. MoP does NOT have a fixed amount of triggers. Purely by the fact that by increasing hit rate, you also decrease C. You also decrease E, increasing DPS. That is why your equation is misleading if not flat out wrong. I shown you in my previous posts where your thinking went wrong and corrected your simplistic equation with steadier variables, accounting for the dynamics of the situation. Unless you can show where in MY equations I am wrong, which CLEARLY shows that increasing hit rate increases DPS, then you have no ground to stand on.

Also, quit with the whining and smack talk. I am not an "idiot."

Quote:

The more minions you have, the faster monsters die. The faster monsters die, the more times you can trigger AP in unit time. The more times you trigger AP in unit time, the lower E is. The lower E is, the higher the DPS is!

Jeydra, stop being a fanboy, and actually read what Cthon is writing. Seriously, I thought better of you.

What you think he's saying as above is exactly NOT what cthon is saying. Cthon is saying you get the same DPS from 100 minions as 1 minion, as long as MoP lasts long enough. Let me quote Cthon:"... so they don't affect DPS except on a granularity smaller than MoP's effective recycle time." Clearly that is wrong - as you have stated, E decreases.

The problem with his thinking is that he is treating E and C as constants.

edit:

Quote:

Imagine two adjacent practice dummies. The time it takes you to autoattack a dummy to death equals one recycle period for proto-MoP. Your damage per auto-attack equals twice proto-MoP's damage per trigger. Imagine casting proto-MoP on one dummy, auto-attacking it to death, thereby taking off half the other dummy's life. Now work out the DPS from proto-MoP over 1 cycle period. Repeat the experiment with a friend also auto-attacking the first dummy. Now work out the DPS from proto-MoP over 1 cycle period. Is it any different? It shouldn't be.

Case 1:
B = dummy HP
A = autoattack DPS
PMOP DPS = damage done / cycle time = (0.5*B)/(B/A) = 0.5A

Case 2:
B = dummy HP
A = autoattack DPS
PMOP DPS = (0.5*B)/[B/(2*A)] = A

In case 2 the DPS is exactly twice case 1. In case 2, the target dummy is killed twice as fast and PMOP triggers twice as fast.

Jeydra · Jul 13, 2009, 10:49 PM // 22:49

Quote:

Originally Posted by traversc

That is patently false. MoP does NOT trigger the same amount of time. Refer to my previous post and below.

MoP does not trigger the same amount of times in unit time, but it certainly triggers the same amount of time before the monster dies. Which is what Chton said, wasn't it?

Quote:

Originally Posted by traversc

What you think he's saying as above is exactly NOT what cthon is saying. Cthon is saying you get the same DPS from 100 minions as 1 minion, as long as MoP lasts long enough. Let me quote Cthon:"... so they don't affect DPS except on a granularity smaller than MoP's effective recycle time." Clearly that is wrong - as you have stated, E decreases.

Of course it isn't what Chton is saying. That's what I wrote above. Read it carefully.

AtomicMew · Jul 14, 2009, 01:41 AM // 01:41

Quote:

Originally Posted by Jeydra

MoP does not trigger the same amount of times in unit time, but it certainly triggers the same amount of time before the monster dies. Which is what Chton said, wasn't it?

As I stated, it does NOT trigger the same amount of time because it decreases C. Physical DPS and non-physical DPS are competing factors. If you increase phys-DPS (say, by increasing the number of minions), you decrease the amount of non-phys damage done per target, increasing the number of MoP triggers per copy.

Quote:

Of course it isn't what Chton is saying. That's what I wrote above. Read it carefully.

Jeydra, I am tired of you saying "just read" "read carefully" "learn to read" etc. especially when you ADMIT to not even reading everything I wrote. It's not helping your argument and it's not welcome.

If you are saying that increasing minion number increases DPS (c.f., "The more minions you have, the faster monsters die. The faster monsters die, the more times you can trigger AP in unit time. The more times you trigger AP in unit time, the lower E is. The lower E is, the higher the DPS is!") then I am in complete agreement, and I am wondering why you bring this point up, considering this is pretty much the basis for my entire argument?

Since you started arguing with me, and not Cthon, I assumed you were in agreement with Chton. If that is not the case, then I apologize. Cthon somehow thinks that more minions does not increase MoP's DPS.

The following questions will clarify for me your opinion, immensely:

Do you, or do you not agree with the statement made by Cthon?:
"Burning through all your triggers in short order can look like a DPS increase ("OMG Barrage just did +140 in 1 sec!!"), but it's really just a spike. Your overall DPS remains constant once you look at a period of time at least as long as a recycle period. Not that spike's aren't nice, but they aren't DPS."

The answer I want from you is a simple yes or no. Considering you realized that killing the target faster decreases E, the recycle period, and increases DPS, I am inclined to think your answer is no.

Awaiting your intelligent reply, Jeydra.

Chthon · Jul 14, 2009, 02:37 AM // 02:37

Quote:

Originally Posted by Jeydra

The more minions you have, the faster monsters die. The faster monsters die, the more times you can trigger AP in unit time. The more times you trigger AP in unit time, the lower E is. The lower E is, the higher the DPS is!

If, per traversc earlier in this thread, you are pointing minions and discord heroes at different targets, the minions should never kill anything fast enough to vary the AP rate in the slightest. For a different sort of team that focuses all its damage on the MoP-ed target, I'd probably have to agree with you; but not for this case.

Quote:

Originally Posted by traversc

Case 2:
B = dummy HP
A = autoattack DPS
PMOP DPS = (0.5*B)/[B/(2*A)] = A

You calculated over 1/2 a cycle period instead of a whole period. Nice try though. Maybe you should use "DPS=total-damage/total-time" instead of what you used -- it makes it harder to make a mistake (or hide an intentional error).

Jeydra · Jul 14, 2009, 04:41 AM // 04:41

@traversc -

Quote:

Originally Posted by traversc

As I stated, it does NOT trigger the same amount of time because it decreases C. Physical DPS and non-physical DPS are competing factors. If you increase phys-DPS (say, by increasing the number of minions), you decrease the amount of non-phys damage done per target, increasing the number of MoP triggers per copy.

I thought Chthon set C to be zero? Read the original quote on post #24.

"One bone minion slowly chewing through a monster 5hp at a time will trigger MoP the same number of times as 50 or100 bone minions chewing through the monster 5hp at a time."

Since all minions do physical damage, there is no non-physical damage coming from the minions, anyway. There's no mention of other sources of damage either. So C = 0. If a monster has 500 HP, then a minion hitting for 5 damage / hit would take a hundred hits to kill it and so trigger MoP 100 times. If there're 100 minions hitting for 5 damage / hit, then the monster would take one hit from every minion and die, and so trigger MoP 100 times. Of course the first case would be slower than the second by a factor of 100, and have a correspondingly lower DPS. But there's no mention of DPS in Chthon's original quote, is there?

Then of course the important thing isn't how many triggers of MoP is but how fast you kill. After all, who cares how many times MoP triggers? The thing to care about is how fast you finish off the mob, and how many times MoP triggers, or AP triggers, or Sundering mods on your weapons trigger, etc. are all just minor facets of building up high DPS. So in this light Chthon was missing the point. Or perhaps he's oversimplifying things. But he isn't wrong.

Quote:

Originally Posted by traversc

If you are saying that increasing minion number increases DPS (c.f., "The more minions you have, the faster monsters die. The faster monsters die, the more times you can trigger AP in unit time. The more times you trigger AP in unit time, the lower E is. The lower E is, the higher the DPS is!") then I am in complete agreement, and I am wondering why you bring this point up, considering this is pretty much the basis for my entire argument?

Because Chton is wrong! Soooooo simple. I mean, just look at what I wrote:

"Btw Chton I didn't read all you and traversc wrote, but this in particular is certainly wrong."

Isn't it obvious that I was talking to Chton not you? Or is English not your first language?

Quote:

Originally Posted by traversc

Do you, or do you not agree with the statement made by Cthon?:
"Burning through all your triggers in short order can look like a DPS increase ("OMG Barrage just did +140 in 1 sec!!"), but it's really just a spike. Your overall DPS remains constant once you look at a period of time at least as long as a recycle period. Not that spike's aren't nice, but they aren't DPS."

The answer I want from you is a simple yes or no. Considering you realized that killing the target faster decreases E, the recycle period, and increases DPS, I am inclined to think your answer is no.

I'm not sure. The sentence is rather hard to understand and the context doesn't help. What's IAtS? I Am The Strongest?

Anyway at first looks it looks to me that first part of this statement I agree with but doesn't really mean anything (see the same about having 1 minion and having 100). That's because DPS is defined to be total damage done over unit time. So "your overall DPS remains constant once you look at a period of time at least as long as a recycle period" is true by definition.

Second part I disagree. Spikes are still DPS. If you deal 500 damage in a second then in that second your DPS is 500, but by their nature spike damage cannot be maintained. You have a downtime. If you deal 500 damage in a second, then 9 seconds without damage (say), then overall your DPS over one cycle is 50. Over a fight though you probably have a higher DPS than this since people tend to die immediately after a spike, not during the 9 seconds of inactivity.

Quote:

Originally Posted by Chthon

If, per traversc earlier in this thread, you are pointing minions and discord heroes at different targets, the minions should never kill anything fast enough to vary the AP rate in the slightest. For a different sort of team that focuses all its damage on the MoP-ed target, I'd probably have to agree with you; but not for this case.

Maybe the minions won't kill anything, but the damage from MoP certainly will. If not, you'd be essentially saying there's little difference in kill speed when you have 20 minions and when you have one. But that's clearly wrong, right?

By the way I'm not aware of any method to point minions at one target and H/H at another. Is there one?

Xenomortis · Jul 14, 2009, 09:16 AM // 09:16

Quote:

Originally Posted by Jeydra

By the way I'm not aware of any method to point minions at one target and H/H at another. Is there one?

Minions will attack the first target that you attack or that enters their aggro range. They will not switch.
Heroes and Henchmen will switch targets to whatever you're attacking (eventually).
I think someone said calling a target will lock them onto it until it's dead or you call a different one.

Improvavel · Jul 14, 2009, 05:56 PM // 17:56

Quote:

Originally Posted by Xenomortis

I think someone said calling a target will lock them onto it until it's dead or you call a different one.

From what I observe, this is true statement. If after you call (ctrl+space), you shift target by just pressing space, the h/h will stay on the called target. If you call a new target, then they will change to your new target.

And that eventually is much more - NOW!

That is one of the reason that makes melee AI so annoying.

AtomicMew · Jul 14, 2009, 06:32 PM // 18:32

Quote:

Originally Posted by Chthon

If, per traversc earlier in this thread, you are pointing minions and discord heroes at different targets, the minions should never kill anything fast enough to vary the AP rate in the slightest. For a different sort of team that focuses all its damage on the MoP-ed target, I'd probably have to agree with you; but not for this case.

Conversely, that simply means that a relatively small difference in C greatly changes the amount of MoP triggers available.

Quote:

You calculated over 1/2 a cycle period instead of a whole period. Nice try though. Maybe you should use "DPS=total-damage/total-time" instead of what you used -- it makes it harder to make a mistake (or hide an intentional error).

Heh. Except AP recharges MoP, so cycle time ~= time it takes to kill a target.

Good game, Cthon.

Quote:

Originally Posted by Jeydra

I thought Chthon set C to be zero? Read the original quote on post #24.

"One bone minion slowly chewing through a monster 5hp at a time will trigger MoP the same number of times as 50 or100 bone minions chewing through the monster 5hp at a time."

The quote does not support your premise. You might think it implies C=0, but it is not explicitly stated.In particular, I believe Cthon to be mistakenly treating C as a constant.

Besides, this entire discussion is about MoP with discordway, and it helps to discuss the general case. But do you agree that C is not constant for the general case?

Quote:

Then of course the important thing isn't how many triggers of MoP is but how fast you kill. After all, who cares how many times MoP triggers? The thing to care about is how fast you finish off the mob, and how many times MoP triggers, or AP triggers, or Sundering mods on your weapons trigger, etc. are all just minor facets of building up high DPS. So in this light Chthon was missing the point. Or perhaps he's oversimplifying things. But he isn't wrong.

No, he was WRONG. Re-read the quote from him in my previous post. He basically states that more minions do not increase the DPS from MoP. ("Burning through all your triggers...") That is flat out wrong, because as you stated, E decreases.

He thinks that result is correct because his equation is overly simplistic. The problem with his equation is that his variables fluctuate wildly and are not independent. I've posted a corrected version of his equation. If you care to read through it, you'll see what I mean. In particular, I've argued that for discordway C is not constant and a small relative variation vastly changes the amount of triggers available. Conversely, this means that a modest increase in minion hit rate will vastly increase MoP DPS. I've also argued that E is not constant, (but I think you argued this more clearly).

PS: When you said "Btw Chton I didn't read all you and traversc wrote, but this in particular is certainly wrong" I thought you were quoting me. Sorry, my mistake! But it wasn't immediately obvious to me, since you also quoted me and that line was under my quote -.-

And yes, english is my first language ...

Improvavel · Jul 14, 2009, 07:34 PM // 19:34

Mark of Pain has limited number of triggers set by the health of the adjacent mobs.

100 minions or 20 minions will triggering exactly the same damage of MoP, for the simple reason the mobs will be death.

Chthon · Jul 15, 2009, 03:03 AM // 03:03

Quote:

Originally Posted by Jeydra

Maybe the minions won't kill anything, but the damage from MoP certainly will.

1. I put the cart before the horse in my previous post. First and foremost, I should have said that AP really has no place in a discussion about MoP's mechanics. AP is something extrinsic to MoP. Taken by itself, MoP's DPS is limited in exactly the manner I described above. Yes, AP creates a feedback loop that can bypass that limitation. But AP does the same thing for every skill with a noticeable recharge, and it does it based on any damage source at all. Citing AP as a reason that MoP (or any other skill) doesn't have limited DPS is kind of like saying that there's no limit on the amount of money you can withdraw from your bank account so long as you bring a gun and point it at the teller.

2. Even if, for argument's sake, we take the recharge boost provided by AP as a legitimate part of the comparison, you're not getting any improved speed from AP in a discordway scenario because of the minions or the MoP -- at least you shouldn't be. You ought to be casting AP on the discord target, which should be neither the MoP focus nor anything next to it. (Also, I might add, you have no hope of casting anything on something adjacent to the MoP focus anyway if you do things right.) Since you're never casting AP on a target that also takes damages from minions or MoP, you're not getting any increase in AP's frequency on account of minions or MoP.

Now, I should say that this problem is limited to non-physical-damage builds like discordway. In a physcial-based team, the MoP-ed target can -- and usually should -- also be the AP target.

Quote:

By the way I'm not aware of any method to point minions at one target and H/H at another. Is there one?

Not reliably, no. I basically said earlier that I don't really believe people who claim they can micro their way out of the discord+MoP anti-synergy. At best you can keep discord and MoP pointed at different targets. But then getting MoP triggered is basically left up to roving minions. But since people insist that they can micro their way out of it, I decided to let it go and stick to the more fundamental point that MoP's DPS is inherently limited.

Quote:

Originally Posted by Improvavel

Mark of Pain has limited number of triggers set by the health of the adjacent mobs.

100 minions or 20 minions will triggering exactly the same damage of MoP, for the simple reason the mobs will be death.

This guy gets it.

(Actually, Improvavel is stating a complimentary limiting factor on MoP's DPS, but is equally true.)

AtomicMew · Jul 15, 2009, 03:53 AM // 03:53

Quote:

Originally Posted by Chthon

1. I put the cart before the horse in my previous post. First and foremost, I should have said that AP really has no place in a discussion about MoP's mechanics. AP is something extrinsic to MoP. Taken by itself, MoP's DPS is limited in exactly the manner I described above. Yes, AP creates a feedback loop that can bypass that limitation. But AP does the same thing for every skill with a noticeable recharge, and it does it based on any damage source at all. Citing AP as a reason that MoP (or any other skill) doesn't have limited DPS is kind of like saying that there's no limit on the amount of money you can withdraw from your bank account so long as you bring a gun and point it at the teller.

Except you originally tried to account for AP's effect. Here, let me jog your memory: "E = average effective cast+recharge time for MoP considering effect of AP"

So now you are saying you are NOT considering the effects of AP?

Please. This entire discussion started when I mentioned playing AP MoP with discordway, which you called "idiocy."

Now it has become clear that you see the point that AP decreases E, just admit you were wrong.

Seriously. Just admit you are wrong. There is far more dishonor in refusing to admit fault than actually admitting fault. Please don't try to finagle the argument by saying "Oh, but I wasn't really considering AP, honestly!"

Quote:

2. Even if, for argument's sake, we take the recharge boost provided by AP as a legitimate part of the comparison, you're not getting any improved speed from AP in a discordway scenario because of the minions or the MoP -- at least you shouldn't be. You ought to be casting AP on the discord target, which should be neither the MoP focus nor anything next to it. (Also, I might add, you have no hope of casting anything on something adjacent to the MoP focus anyway if you do things right.) Since you're never casting AP on a target that also takes damages from minions or MoP, you're not getting any increase in AP's frequency on account of minions or MoP.

Quote:

Improvavel is stating a complimentary limiting factor on MoP's DPS, but is equally true.

Yes, improvel is correct, but it is not in support of your argument. And I don't think he saying it is a limiting factor on DPS, because, believe it or not, mobs dying faster means higher DPS.

the slasher · Jul 15, 2009, 11:19 AM // 11:19

I'm sorry, is this off-topic? Yes, the E/N build first suggested does use MoP, and you coooould say you're discussing if the build is better/worse (its obvious by now) than the regular N/A MoP caster, but.... if you want to discuss how MoP works on general PvE, I suggest you go to another thread... need an admin here...

and BTW, chnton, DPS= damage per second....

assuming you have your Dummies you talked about... you have an AP caller using MoP, meaning when the MoP'ed dummie dies, you have your MoP recharged...

now, tell me, you have 30 minions from 3 minion masters (lets assume you are using 2 man discord and took 3 mm's)... all your party members, just for this test, stop attacking, and you only cast MoP+AP on the target dummie

lets see how they fare...

20 minions get to the dummie, each one attacks once, dummie is dead (till here, everything ok?)
MoP dmg is for certain HIGHER than minions average dmg, but just in case, lets say, how you stated be4, just for the test, lets say MoP dmg per trigger = 1/2 of auto attack dmg * 2 (2 adjacent dummies) (which equals auto attack dmg)

well... with 1 minon attacking the dummie, you would kill the MoP dummie in 20 hits (lets say MoP is long enough for that to happen while always triggering MoP

now, about DPS, which is what everyone talks about...

whe n 20 minions attack, you do damage=auto attack minion damage*20+(MoP trigger dmg*20*2) in one auto attack (which, lets say, is 1 second(minion auto atack rate is way lower though, getting to 3.3 seconds or so)..

with 1 minion, you do get the same DAMAGE.. but not the same DPS.. thers a difference.. overall dmg, is the same, dps not... (i hope its clear why), ... +, you will have MoP recharged faster(ap), meaning you can MoP another target to dead. anyhow, the formula for the DPS (dmg per second ) would be(for one target to die) Damage= minion auto attack Dmg* number of minions(20 in this case)/number of hits each minion must attack to kill target (which in this case is 1)+(MoP trigger damage*number of hits till target is dead(20 in this case)*number of adjacent foes(2 in this case))

you now realise what you've done wrong? with 1 auto attack from a minion or w/e, you must do /20 in the dps.... not for overall dmg though

now, this has been stated be4, just thought abit unclearly... (oh, btw, i know the numbers for attack rate and dmg are wrong, but the formulas are correct

speedy21589 · Jul 15, 2009, 01:40 PM // 13:40

Let's put it this way...from what I've read, it seems Chthon is talking about average DPS, whereas traversc is talking about "instantaneous" DPS.

An analogy: speed. If we're driving from point A to point B, the average speed is total distance traveled divided by total time spent. The instantaneous speed at any point during this trip, however, could be anything. If for the whole trip we went 60 miles in 60 minutes, there's nothing to say that we didn't travel at a constant instantaneous speed of 3000 mph for 1 minute (that would be 50 miles traveled in that minute) and covered the last 10 miles over 59 minutes (essentially 10 mph for those 59 minutes).

What we're dealing with here is even more extreme than that. Let's drop AP for now for simplicity. I believe that Chthon is arguing that the damage MoP can do in its time up is limited by the life of adjacent foes. Dividing this maximum damage by the total recharge time of MoP gives the average DPS of one MoP. In other words, point A is when the first MoP is cast, and point B is when it has recharged.

However, all adjacent foes could be blown up in the first 2 seconds with enough triggers in those 2 seconds. That means that, in those first 2 seconds, MoP is doing a much higher instantaneous DPS than the average DPS done over the whole 21 seconds between point A and B, while over the rest of the time, the instantaneous DPS is 0. It would be like covering those 60 miles in the first 2 minutes and then sitting around for 58 minutes, then calling your average speed for the whole trip 60 mph. Not technically "wrong," because you are allowed to define an average that way...but does that number mean anything?

So I think the question is, is average DPS or instantaneous DPS a more useful measurement?

bsjonson · Jul 15, 2009, 02:00 PM // 14:00

Quote:

So I think the question is, is average DPS or instantaneous DPS a more useful measurement?

An interesting point, which I think can be taken one logical step further...which is a more realistic example for actual gameplay?

IMO (feel free to flame me if I'm wrong), the instantaneous DPS is more useful in a build like AP + minions because you are spamming Mark of Pain & then overloading the mob with 10-15 triggers of MoP (assuming 60-70% of the minions get to the target & each hit 2-3 times which is a fairly realistic expectation), following which you will be cleaning up (ideally with barbs or just YMLAD + AP spam).

The question then becomes, how fast are you going to kill the called target? If you are running a ton of physicals, even in HM, an enemy wont usually stay up long enough to get 2-3 Dragon Slashes off, or a Death Blossom Moebius chain for that matter...so then it starts to become comparable to the AP Discord scenario.

If you are running some weird pressure build where each kill takes you 15-20 seconds...then yes, average DPS is relevant. But if not, I feel instantaneous DPS is more realistic as that is what MoP really is all about...a barrage of 42s all over your screen, following which enemies scatter and then you go in on clean up duty.

maxxfury · Jul 15, 2009, 04:52 PM // 16:52

What do you ideally want to happen?

you want Mop to proc as many times as possible in the shortest space of time! to blow the mop up as fast as possible!

Correct? yes, to a point.

The limit to how many procs you need is the health of the mob.

So having 30 minions + casters with spears + evas's + your melee's will make very little difference to having 10 minions +casters with spears + evas's + your melee's.

As the Mop will proc enough to kill the mob before all 30 minions get a chance to hit! So overkill..Hell it powerful enough to not need ANY minions to blow up a full group! just your caster spears + sins (+your phys if you bring any!) IF you play a mop nuker correctly.

If you DONT play it right, minions will help, as your slacking on getting your procs and killing to slow.

More minions give you more room for error, people not following calls, stupid ai, the usual fail excuses, a big "trash wall" and the usual additional damage from the minions hitting what ever they hit.

If its just an extra hex in a necro disco caller? why not..wont hurt. wont get huge amounts out of it, compared to a tailor made mop-nuke set up. even if you micro it on another target and spike another. will it do damage? yes, will it matter too much either way? no.

AtomicMew · Jul 15, 2009, 04:52 PM // 16:52

Quote:

Originally Posted by speedy21589

What we're dealing with here is even more extreme than that. Let's drop AP for now for simplicity. I believe that Chthon is arguing that the damage MoP can do in its time up is limited by the life of adjacent foes. Dividing this maximum damage by the total recharge time of MoP gives the average DPS of one MoP. In other words, point A is when the first MoP is cast, and point B is when it has recharged.

Bolded part is wrong. You can't drop AP as that defeats the purpose. In context, if you care to read through the last two pages of TLDR, you'll see the discussion veered off to AP MoP with discordway.

@ The slasher: Spot on.

speedy21589 · Jul 15, 2009, 05:17 PM // 17:17

Quote:

Originally Posted by traversc

Bolded part is wrong. You can't drop AP as that defeats the purpose.

I was dropping AP temporarily to help illustrate that it seems like you and Chthon are looking at DPS in different ways. He has defined DPS in a long-term sense, whereas you are more concerned with instantaneous DPS.

With or without AP, MoP has the potential for near infinite instantaneous DPS. Without AP, this means you can blow up adjacent foes quickly, then sit around waiting for it to recharge. With AP, this means you can blow up adjacent foes quickly, then blow up more adjacent foes again, and again, without being as limited by the recharge.

Jeydra · Jul 16, 2009, 02:49 AM // 02:49

Quote:

Originally Posted by traversc

Besides, this entire discussion is about MoP with discordway, and it helps to discuss the general case. But do you agree that C is not constant for the general case?

There's no doubt that when I have 20 minions as opposed to 1 minion about me, I deal damage a lot faster. Just how much faster, I do not know. But I know mobs die faster and I pull off AP triggers a lot more often.

That said, I have not done MoP + AP before, so I can't say for certain. Let's just say it seems very likely that C is not constant.

Quote:

Originally Posted by Chthon

1. I put the cart before the horse in my previous post. First and foremost, I should have said that AP really has no place in a discussion about MoP's mechanics. AP is something extrinsic to MoP. Taken by itself, MoP's DPS is limited in exactly the manner I described above. Yes, AP creates a feedback loop that can bypass that limitation. But AP does the same thing for every skill with a noticeable recharge, and it does it based on any damage source at all. Citing AP as a reason that MoP (or any other skill) doesn't have limited DPS is kind of like saying that there's no limit on the amount of money you can withdraw from your bank account so long as you bring a gun and point it at the teller.

2. Even if, for argument's sake, we take the recharge boost provided by AP as a legitimate part of the comparison, you're not getting any improved speed from AP in a discordway scenario because of the minions or the MoP -- at least you shouldn't be. You ought to be casting AP on the discord target, which should be neither the MoP focus nor anything next to it. (Also, I might add, you have no hope of casting anything on something adjacent to the MoP focus anyway if you do things right.) Since you're never casting AP on a target that also takes damages from minions or MoP, you're not getting any increase in AP's frequency on account of minions or MoP.

I think you should certainly consider AP's effects, because AP makes and breaks bars. Skills like YMLAD!, Finish Him! and the Vanguard Assassin are all really strong, but only because of AP. Without AP, their long recharge drags them down. I think MoP falls into this category as well. Without something to lower the recharge on it, it's a lot weaker.

By the way I usually reserve AP till someone is near-death, usually with sufficiently low HP to finish off with the two Norn shouts or at most a bit of prodding / calling / Lightning Orb. This protects against missed AP's and hex removal. I don't know what works best N/A; I don't have a PvE Necro after all. Still it seems to me that if you point minions one way and AP someone else, you'll still get faster recharges (via AP) with 20 minions than 1. That's because the minions hit the target with MoP, MoP triggers and everyone else around it takes damage, and that damage means you trigger AP sooner. You write that you have no hope to cast AP on someone near the MoP'ed target. Care to elaborate? That's like saying MoP does no damage because there won't be anyone near the marked target. Sounds pretty silly to be honest.

Quote:

Originally Posted by bsjonson

IMO (feel free to flame me if I'm wrong), the instantaneous DPS is more useful in a build like AP + minions because you are spamming Mark of Pain & then overloading the mob with 10-15 triggers of MoP (assuming 60-70% of the minions get to the target & each hit 2-3 times which is a fairly realistic expectation), following which you will be cleaning up (ideally with barbs or just YMLAD + AP spam).

I think average DPS is the only thing that matters. If you cannot keep up spike DPS, reporting that as your DPS is just wrong ... I can for example deal 1600 DPS as an Elementalist primary by hitting YMLAD! and Finish Him! within 0.1 seconds of each other, except it doesn't work that way right?

If you spam enough MoP's / Death Blossoms / Dragon Slashes and overload the mob in 10 seconds, then I'd say your average DPS is very high. It's not instantaneous DPS however. Unlike the above, you can keep up the damage, for as long as you need to.