test content
Logo
What is the Arc Client?
Install Arc
Testing At-Will Powers
pfft2
Member Posts: 301 Arc User
I've wondered for awhile how exactly the various CW At-Will powers compare quantitavely, so when I saw yesterday that the new patch fixed Storm Pillar's (in)ability to crit on the first strike, I decided to head over to the test server and fire up ACT to see what I could find.
[UPDATE 7/30/2013: I've reorganized the thread to account for future testing. The first post will contain an overview of results, with links to the relevant data. My first tests -- describing the basic functionality of each power, and now including visual effect gifs! -- have been moved to the second and third posts of the thread. My old disclaimer still applies, however. If you wish to argue the practical merits of this-or-that At-Will power, or if you wish to dispute my results, then by all means, feel free to use this thread for that purpose -- but do understand that I only wish to explore the mechanics as objectively as I can; my tests are not a veiled attempt to advocate any particular power or build.]
Methodology:
Basic Functionality and Baseline Testing Summary (7/19/2013):
Links: Magic Missile, Chilling Cloud, and Ray of Frost results, Storm Pillar results
Class Features:
[UPDATE 7/30/2013: I've reorganized the thread to account for future testing. The first post will contain an overview of results, with links to the relevant data. My first tests -- describing the basic functionality of each power, and now including visual effect gifs! -- have been moved to the second and third posts of the thread. My old disclaimer still applies, however. If you wish to argue the practical merits of this-or-that At-Will power, or if you wish to dispute my results, then by all means, feel free to use this thread for that purpose -- but do understand that I only wish to explore the mechanics as objectively as I can; my tests are not a veiled attempt to advocate any particular power or build.]
Methodology:
Unless otherwise noted, all tests were conducted on my level 60 CW, wearing nothing at all except her weapon (a Burning Shard). I also respecced and left all of my feat points and ability-score points unspent. I ensured I had no campfire bonus or any other temporary buff. My character's stats in that blank-slate state follow:
INT/WIS/CHA - 20/13/13
Crit Chance - 9.8%
Crit Severity - 75%
Power - 1,569
Weapon Damage - 641-784
Each At-Will power was given max ranks, and each was spammed for a minute or more at a single dummy in the Trade of Blades. Neither Chilling Cloud nor Storm Pillar ever hit more than one target that I could see. (The lingering Storm Pillar effect did occasionally hit an adjacent dummy, but as far as I can tell that makes no difference; the pillar's attacks are all single-target, and there's a hard limit on how many attacks it'll make during its short lifespan, so all damage counts as if it were against a single target for our purposes. This quirk in Storm Pillar may affect its practical usability in niche situations, but it doesn't appear to affect our evaluation of its damage potential.)
INT/WIS/CHA - 20/13/13
Crit Chance - 9.8%
Crit Severity - 75%
Power - 1,569
Weapon Damage - 641-784
Each At-Will power was given max ranks, and each was spammed for a minute or more at a single dummy in the Trade of Blades. Neither Chilling Cloud nor Storm Pillar ever hit more than one target that I could see. (The lingering Storm Pillar effect did occasionally hit an adjacent dummy, but as far as I can tell that makes no difference; the pillar's attacks are all single-target, and there's a hard limit on how many attacks it'll make during its short lifespan, so all damage counts as if it were against a single target for our purposes. This quirk in Storm Pillar may affect its practical usability in niche situations, but it doesn't appear to affect our evaluation of its damage potential.)
Basic Functionality and Baseline Testing Summary (7/19/2013):
Links: Magic Missile, Chilling Cloud, and Ray of Frost results, Storm Pillar results
Magic Missile -
Attacks per second: 694 / 312 = 2.22
Time for a full "round" of 5 attacks: (312/694) * 5 = ~2.25 seconds (11.25 seconds to maximize Arcane Mastery stacks)
DPS: 1292.16
Crit-neutral DPS: 1292.16 / (1 + (0.10 * 0.75)) = 1202
Approximate DPS without crits and without Arcane Mastery: 1202 / 1.15 = 1,045.2
Time for a full "round" of 5 attacks: (312/694) * 5 = ~2.25 seconds (11.25 seconds to maximize Arcane Mastery stacks)
DPS: 1292.16
Crit-neutral DPS: 1292.16 / (1 + (0.10 * 0.75)) = 1202
Approximate DPS without crits and without Arcane Mastery: 1202 / 1.15 = 1,045.2
Chilling Cloud -
Attacks per second: 111/84 = 1.32
Time for a full "round" of 3 attacks: (84/111) * 3 = ~2.27 seconds
DPS: 841.51
Crit-neutral DPS: 841.51 / (1 + (0.11 * 0.75)) = 777.37
Time for a full "round" of 3 attacks: (84/111) * 3 = ~2.27 seconds
DPS: 841.51
Crit-neutral DPS: 841.51 / (1 + (0.11 * 0.75)) = 777.37
Ray of Frost -
Attacks per second: 192/113 = ~1.7
DPS: 642.79
Crit-neutral DPS: 642.79 / (1 + (0.11 * 0.75)) = 593.8
Time to max chill stacks (at 6) on target: ~3.5 seconds.
DPS: 642.79
Crit-neutral DPS: 642.79 / (1 + (0.11 * 0.75)) = 593.8
Time to max chill stacks (at 6) on target: ~3.5 seconds.
Storm Pillar -
Attacks per second (main attack, used for procing purposes): 41/96 = 0.427
Attacks per second (total): 245/96 = 2.55
Time for a full "round"/charge of attacks: 96/41 = 2.34 seconds
DPS (all attacks): 140765/96 = 1466.3
Crit-neutral DPS: 1466.3 / (1 + (0.1 * 0.75)) = 1364
Attacks per second (total): 245/96 = 2.55
Time for a full "round"/charge of attacks: 96/41 = 2.34 seconds
DPS (all attacks): 140765/96 = 1466.3
Crit-neutral DPS: 1466.3 / (1 + (0.1 * 0.75)) = 1364
Class Features:
Storm Spell Summary (7/30/2013, post Fury of the Feywild patch) - (Click here for full results)
Magic Missile -
Total number of MM attacks: 861
Total number of Storm Spell procs: 108
Storm Spell proc rate: 108 / 861 = 12.5%
Total number of Storm Spell procs: 108
Storm Spell proc rate: 108 / 861 = 12.5%
Chilling Cloud -
Total Number of CC attacks: 1,212
Total Number of Storm Spell procs: 344
Storm Spell proc rate: 344/1212 = 0.2838, or ~28.4%
Total Number of Storm Spell procs: 344
Storm Spell proc rate: 344/1212 = 0.2838, or ~28.4%
Ray of Frost -
Total Number of RoF attacks: 1,127
Total Number of Storm Spell procs: 376
Storm Spell proc rate: 376/1127 = 0.3336, or about 33.4%
Total Number of Storm Spell procs: 376
Storm Spell proc rate: 376/1127 = 0.3336, or about 33.4%
Storm Pillar -
Total number of Storm Pillar attacks: 1000 + 5000 = 6,000
Total number of Storm Pillar procs: 172
Storm Spell proc rate: 172 / 6000 = 2.86%
Total number of Storm Pillar procs: 172
Storm Spell proc rate: 172 / 6000 = 2.86%
Post edited by pfft2 on
0
Comments
[NOTE: The data and power descriptions that appear in these basic-testing posts are not intended to comprise a holistic and practical comparison of the various At-Will powers. The first two posts describe each power's base functionality as of July 19, 2013, nothing more. I will occasionally update the thread with new testing to describe feat and class feature interactions.]
Magic Missile:
Tooltip damage under the tested conditions: 428-504
My first test of MM was shorter than I thought; the parse showed only 55 seconds of attacking. And the parse showed a very low crit rate. To verify my results (and to straighten out the crit result), I ran it again for about 5 minutes. Results follow:
Magic Missile Log
Time: 312 seconds
Damage total: 404,188
Hits: 694
Attacks per second: 694 / 312 = 2.22
Time for a full "round" of 5 attacks: (312/694) * 5 = ~2.25 seconds (11.25 seconds to maximize Arcane Mastery stacks)
DPS: 1292.16
Crit-neutral DPS: 1292.16 / (1 + (0.10 * 0.75)) = 1202
Approximate DPS without crits and without Arcane Mastery: 1202 / 1.15 = 1,045.2
Magic Missile is the only Control Wizard At-Will power that comes packaged, by default, with its own damage-increasing buff. Arcane Mastery's damage bonus is an intrinsic advantage of MM, and I don't mean to take that away; I only list MM's (approximate) DPS with AM factored out for comparison purposes. (And to let people know that they doesn't start at 1202 DPS.)
Apart from Storm Pillar, which has a proc-rate disadvantage we'll describe below, MM is the CW's fastest single-target attack. It might therefore be the best choice if you're looking to proc weapon enchants or Storm Spell, but I need to test further to be sure: an early test showed that Ray of Frost has an insanely high proc rate with Storm Spell even despite RoF's slower attack rate. Not sure whether that was just a fluke or a bug or an intended design quirk. Will have to look further into it.
Chilling Cloud:
(Note: the Chilling Cloud gif here shows the new behavior of Chilling Cloud's AoE as of July 30th on the Test server -- in other words, after the Fury of the Feywild patch. The test results in this post are, however, from before that patch. I took the footage for the gif later, obviously. More info on the changes to Chilling Cloud is forthcoming.)
Tooltip damage under the tested conditions: 343-403
Chilling Cloud Log
Time: 84 seconds
Damage total: 70,771
Hits: 111
Attacks per second: 111/84 = 1.32
Time for a full "round" of 3 attacks: (84/111) * 3 = ~2.27 seconds
DPS: 841.51
Crit-neutral DPS: 841.51 / (1 + (0.11 * 0.75)) = 777.37
Chilling Cloud's the go-to AoE At-Will power for chill-based builds. For those who aren't familiar with the power, it has a three-attack cycle, wherein the first two attacks are single-target strikes hitting for middling (tooltip-ish) damage, and the third hitting in a small area of effect for roughly double the damage of each of the previous two.
As the name implies, Chilling Cloud applies the Chill effect, but only on the third strike. A recent(ish) patch gave Chilling Cloud's first two attacks the ability to maintain previously applied stacks of chill, but if you're just using Chilling Cloud, it'll take about 13.62 seconds to max out chill stacks (6) on your target(s). Also, Chilling Cloud won't freeze anything by itself; you need to combine its chilling effect with a specific freezing power (Icy Terrain, Ray of Frost) to get that effect.
That said, there are a number of feats and Class Features that increase Chilling Cloud's damage. I hope to use this thread as a place to compare the aggregate damage output of the various At-Wills, given varying builds and situations. For now, though, we'll stick with the basics.
Ray of Frost:
Tooltip damage under the tested conditions: 321-378
Ray of Frost Log
Time: 113 seconds
Damage total: 73,085
Hits: 192
Attacks per second: 192/113 = ~1.7
DPS: 642.79
Crit-neutral DPS: 642.79 / (1 + (0.11 * 0.75)) = 593.8
Time to max chill stacks (at 6) on target: ~3.5 seconds.
Ray of Frost is, without question, the most defensive/controller-y At-Will we have. As you can see, the damage is rather low, but the power has a great deal of utility in single-target situations (particularly in PvP). Every tick of the Ray applies a stack of chill, and when the target reaches 6 stacks, he freezes solid.
For weeks, I've heard that Ray of Frost procs Storm Spell better than an other At-Will. I was skeptical because to my untrained eye it looked like MM attacked significantly faster than MM. Testing proves that I was right about Ray's attack speed relative to MM, but late last night, it sure looked like Ray has a vastly increased proc rate. I'm not 100% sure it wasn't a mirage or a fluke, but we'll come back to that another time.
Like MM and Chilling Cloud, Ray can benefit from a pretty wide variety of feats and class features.
(Continued)
Storm Pillar:
Tooltip damage under tested conditions: 1269-1507
Storm Pillar's an odd duck. When I first tried it out at level 35 (or whenever it first becomes available), I found the power deeply underwhelming: you have to stand in place to charge it, and your only reward for that effort is a mediocre burst of damage (in a small AoE, similar in size to the AoE on Chilling Cloud's third strike) and a pseudo-pet effect (a pillar of light) that lingers for about 1 seconds (at rank 1) afterwards, doling out a tiny handful of small single-target damage ticks.
The first strike of the power, as I understood it, also couldn't crit. And the selection of feats (and class features) that improve Storm Pillar is ... terrible, nearly non-existent. And I found that lack of lightning-themed feat selections morbidly amusing, because the Paragon Feats are supposed to be based on our Paragon Path, are they not? And our only Paragon Path is the very same Paragon Path that gives us Storm Pillar in the first place.
You know, the Paragon Path called, "Spellstorm Mage?!?"
Anyway, after my brief first experience with Storm Pillar, I threw it away in disgust, leaving it to languish, neglected, in my powers' list at one rank. But yesterday, I found myself experimenting again. With three points in the power, and with its shiny new ability to crit, Storm Pillar suddenly seemed pretty **** good! Since it doesn't have a beam or a projectile (and therefore doesn't telegraph your location), Storm Spell's an ideal power to use in ambush situations (like when you're sitting on top of one of the columns in the Hotenow Domination PvP map).
And the charge time doesn't seem all that long anymore! And at rank three, the pillar lasts for a good two seconds, and tosses out like 6 ticks of Magic-Missile-comparable damage! I thought we might have found a keeper. My results?
Storm Pillar Log
There are two components to Storm Pillar's parsed damage. The first is a big(ish) damage attack that can hit multiple targets in a small area (a la Chilling Cloud). The next six or so attacks come from the pillar effect, which ACT considers a separate entity. The upside? The total damage on the pillar's attacks is pretty impressive, and yes, the pillar's attacks can crit. The downsides are that the pillar's attacks don't appear to proc Storm Spell, and you have no control over what they hit -- so if you're in a situation where one particular mob needs to die now, but he also has a crowd of friends around him, Storm Pillar won't necessarily give you the full benefit of its single-target potential on the particular target you have in mind.
Time: 96 seconds
Damage total: 64,666 + 76099 = 140,765
Hits (main attack): 41
Hits (pillar): 204
Hits (total): 245
Attacks per second (main attack, used for procing purposes): 41/96 = 0.427
Attacks per second (total): 245/96 = 2.55
Time for a full "round"/charge of attacks: 96/41 = 2.34 seconds
DPS (all attacks): 140765/96 = 1466.3
Crit-neutral DPS: 1466.3 / (1 + (0.1 * 0.75)) = 1364
As you can see, Storm Pillar at rank 3 has the highest default DPS of any CW At-Will power. It even beats Magic Missile at 5 stacks of Arcane Mastery, and even against a single target (provided, as above, that you aren't too concerned about having a particular mob in the midst of a group die first). Unfortunately, Storm Pillar loses ground in the proc-rate race, coming in at a lusty dead last, and there is (off the top of my head) only one feat that offers you a significant gain in terms of damage for using it. (Destructive Wizardry in the Thaumaturge tree offers a 10% damage buff for hitting more than one target with the first strike of Storm Pillar.)
Will hope to do more later.
Ray of Frost has synergy with storm spell feature and do equal damage with missles when feated properly.
Chilling cloud must be feated too.
Not without help from feats or class features. Storm Pillar has higher DPS even if you assume full Arcane Stacks for Magic Missile. Granted, Magic Missile will almost certainly edge out Storm Pillar in single-target damage once feats and features (storm spell) are factored back in, but the base numbers are worth knowing.
The first claim isn't obviously true; we'll see whether RoF can equal MM, but I seriously doubt it. Ray of Frost's crit-neutral DPS is 593.8. Magic Missile's crit-neutral DPS without Arcane Mastery stacks is ~1,045. That means you'd have to increase RoF's damage by ~76% before you even reach Magic Missile's baseline DPS. You'd have to increase RoF's damage by ~102% (1202/593.8)-1) to reach MM's DPS with just Arcane Mastery.
And Magic Missile has a few feats up its sleeve too. Frankly, RoF shouldn't have MM-comparable damage, given that RoF can freeze targets. MM's your basic damage tool; RoF's a utility/control tool that also happens to do decent damage.
And yes, I agree with your second statement. As noted in the OP, Storm Pillar lags the others with respect to the selection of useful feats and class features that boost its damage (and/or increase its general utility).
Can you rerun this with Storm Spell included in the total? That's the only way to actually measure comparable damage output.
However, I am interested in Storm Pillar, as an opener (with Destructive Wizardry).
This is just the first step. I wanted to establish a baseline before venturing off into all of the complications.
With Storm Spell, you need an extremely long test -- maybe ten minutes or more of straight attacking -- because if there is a hidden proc-rate bonus on (for example) Ray of Frost, we have to rule out any flukiness. FWIW, the first, relatively short test I logged with RoF and Storm Spell yielded a proc rate of something like 32%. Needless to say, that's a lot higher than the 10% you'd expect. By that point, though, it was late at night and all I could do was stare dumbly at that result; I have no idea if it's accurate.
We could afford to do shorter tests with the base powers because we have a reasonable expectation that crit rate is the same for all powers, and thus we could factor out any small irregularities in the crit rate. (In this case, my character sheet showed a crit rate of 9.8% and all of the tests, except for one, showed 10-11%. When I saw a lower-than-expected crit rate in my first, shorter test with magic missile, I reran MM for five minutes and saw the crit rate normalize.) That's not to say that I think the tests above are iron-clad; I'm sure there are irregularities, but the main thing was simply to iron out the relationship between damage and cast time, because I've never seen the latter quantified.
Anyway, the long and the short of it is that I need to do more testing. Along with Storm Spell, I'd like to iron out exactly how the various feat bonuses interact with the various powers. I'm pretty sure most of them work as you'd expect, but there's only one way to be sure.
Nor is it meant to be. This is about quantification, not recommendation. I suspect that we'll ultimately discover that each power has its role, that none objectively out-performs the others unreservedly, but as of this moment all I can say is that I'm pleasantly surprised by how well Storm Pillar performs out of the box (at rank 3).
For a long time, I've done the same thing, largely on the advice of your guide. Still, I'm not 100% sure what you mean when you say that Chilling Cloud does "95% of Magic Missile damage on [a] single target." Does your Chilling Cloud perform at 95% of Magic Missile's baseline level? Its baseline level with full Arcane Mastery stacks and nothing else? Its best-possible damage given the fullest possible feat investment in enhancing Magic Missile?
That's what this thread is about, answering those questions.
In that case, get to work.
It is comparing Magic Missile to Chilling Cloud specifically with the spec that I am running. So no it is not a absolute side by side comparison. It only takes into account the synergy I am able to create using my current spec. It wasn't an extensive test either, but here is the general output.
Chilling Cloud
Chilling Cloud
Chilling Cloud
Chilling Cloud
Magic Missile
1x Target
2x Target
3x Target
1x Target
Base 1st hit
491
0
0
618
Base Full Rnd
2583
3955
5327
3057
5x Arcane Mastery (15%)
2583
3955
5327
3516
Arcane Enhancement (6%)
2583
3955
5327
3726
Blighting Power (9%)
2815
4311
5806
3726
Bitter Cold (5%)
2956
4526
6097
3726
Froz. Pow Trans (5%/perT)
3104
4979
7011
3726
Chilling Presence (18%)
3663
5875
8273
3726
Thanks. I've put a note at the beginning of the first post to clarify my purpose here.
Good stuff, thanks. I think I've seen that before now that you mention it. And now I have confirmation that the cast times of both powers are identical, which jives with my subjective experience, but I wanted to know for sure. (I've measured MM's attack cycle at ~2.25 seconds; CC's attack cycle takes ~2.27. I suspect they're designed to be exactly the same.)
What prompted my testing, perhaps more than anything else, was the claim (made several times on the forums, not by you in particular) that Ray of Frost attacks faster than MM. I didn't feel that was true, and now I know it isn't true. MM is faster than RoF.
But -- and this is a big "but" -- it may very well be that RoF gives you more procs even despite its lower attack speed. So maybe everyone's right after all, though perhaps in different ways than they'd imagined.
Yeah, the chill-stacks-proccing-Storm-Spell thing was my first suspicion. If it's true that Storm Spell procs every time a chill stack is applied, then that might influence our evaluation of powers like Chilling Cloud and Icy Terrain too. It might make the biggest difference for Conduit of Ice, which doesn't apply Chill unless it's slotted in the Mastery slot.
Or maybe it's just a quirk with Ray of Frost's chill stacks. Either way, this whole line of reasoning raises questions about the Glacial Movement feat (Oppressor); if it boosts the proc rate on Ray of Frost by 25%, then that's a not-insignificant damage bonus packaged with a feat that we've traditionally considered control-themed.
Methodology:
Magic Missile -
Anyway, you'll find a copy of my combat log here. The results follow:
Our Magic Missile tooltip is identical, post-patch, to the tooltip two weeks ago, listing a damage range of 428-504 damage.
And our parse results are likewise consistent with our previous tests: Magic Missile shows a DPS of 1,282.04, including crits and Arcane Mastery. In our previous test, DPS was at 1,292.12, so our new results are within 1% of our old ones. Always encouraging!
But what we're really interested in here is Storm Spell's proc rate:
Total number of Storm Spell procs: 108
Storm Spell proc rate: 108 / 861 = 12.5%
That's close enough to the expected value of 10% that I could just have a bit of a flukey streak. The Magic Missile test, after all, wasn't as long as I'd have liked. May have to come back and test again. The proc rate on some other abilities can't be hand-waved so easily, though, as we'll see.
Storm Spell's overall contribution to damage output: 81992 / 583013 = 14%, or a 583013 / 501021 = 16.3% boost over Magic Missile, alone.
Chilling Cloud -
In fact, Chilling Cloud's tooltip also remains identical, two weeks after our first test, reporting a damage range of 343-403. Combat Log link here.
Happily, our results are once again consistent. Our post-patch 839.77 DPS is within 0.3% of our previous value. So at least with regard to single-target damage (and without any feats), Chilling Cloud remains the same.
As to Storm Spell's proc rate with Chilling Cloud?
Total Number of Storm Spell procs: 344
Storm Spell proc rate: 344/1212 = 0.2838, or ~28.4%
It appears that cold-based spells do have a sizable advantage when it comes to Storm Spell procs, but what's interesting here is that there doesn't appear to be a direct correlation between the frequency of chill applications and the proc rate of Storm Spell; after all, Chilling Cloud only applies chill once per three attacks, and yet CC also boasts a Storm Spell proc rate nearly three times higher than the tooltip would have you believe. And if Chill were the explanation for the increased proc rate, you'd expect Ray of Frost to proc 2 or 3 times as much as Chilling Cloud, but as you'll see, it doesn't.
And since this was a lengthy parse (more than 15 minutes, with well over 1,000 attacks logged), and since the proc rate number is so much higher than 10%, I'm far less inclined to chalk this one up to flukiness.
(And no, I didn't hit multiple targets with Chilling Cloud in this test. Given that post-patch CC has a much larger area of effect, it's harder to find a dummy that's isolated enough to keep CC single-target-only, but it's not impossible to do either.)
Anyway, in this test, Storm Spell accounts for a whopping ~26% of the total damage. Or in other words, slotting Storm Spell gives you a 1043435 / 775441 = 34.5% damage boost over Chilling Cloud by itself (sans other feature/feat bonuses, that is).
Ray of Frost -
Like Magic Missile and Chilling Cloud, Ray of Frost remains unchanged for the purpose of this test. The post-patch tooltip is likewise identical to our tooltip from two weeks ago, reporting a damage range of 321-378 on our naked mage.
Combat Log link here.
Once again, our post-patch test results are consistent with our results from two weeks ago, though there is a little more variance here than we saw with Magic Missile and Chilling Cloud. The new DPS number, 657.24, is ~2.2% higher than our old number, 642.79. Still, the results are close enough that I'm comfortable declaring the power's damage unchanged.
On to the nitty gritty; is Ray of Frost the undisputed champion of the Storm-Spell proc-rate derby? Why yes, yes it is:
Total Number of Storm Spell procs: 376
Storm Spell proc rate: 376/1127 = 0.3336, or about 33.4%
As noted above, Ray of Frost's proc-rate advantage is far too small if the explanation is chill. After all, RoF applies a chill stack once per ~0.6 seconds, whereas Chilling Cloud applies chill once per ~2.25 seconds. There has to be another explanation given that the two powers' proc rates are only ~5% apart, and given that Chilling Cloud procs nearly three times more often than the tooltip suggests.
Anyway, in this case, Storm Spell accounts for 41% of the total damage. It represents an astounding 70% damage boost over Ray of Frost by itself (and sans feat/feature bonuses)!
Storm Pillar -
Sooooo ... if you want to log let's say 1,000 attacks to pin down the incidence of a random proc effect, you need 2.3 * 1000 = 2300 seconds of loggage. That's more than 38 minutes. I got it done, but I had to teach myself how to make a macro to perform the task 1,000 with more-or-less perfect timing first. And then I had to sit there watching like a hawk lest some patch-curious passer-by come along and mess with my target dummy. Happily, no one did.
That said, keep in mind that I arrived at the 2.3-second figure by human testing, so my using that just-slightly-flawed number in a macro doesn't give this parse an unrealistic level of precision; I figure there's just enough error in there to give us a fair approximation of what a reasonably good player can expect, on average, if he spams the power over any combat-realistic period of time.
I also wanted to give Storm Pillar a little extra log time because, unlike all of the other At-Wills, Storm Pillar's tooltip did change in the new patch. Before the tooltip reported a damage range of 1269-1507. The post-patch tooltip reports a damage range of 1348-1586, or about a 5.7% increase in average damage. Is that simply a reporting error corrected, or is it an albeit small buff to the power's real damage? We shall see.
On to the results. You'll find the log here.
As before, there are two components to Storm Pillar's damage. There's the main strike, which hits in a small AoE, and then there's a lingering pseudo-pet effect that parses as a separate entity. If we combine the DPS of the main strike with the DPS of the Pillar (ignoring the Storm Spell proc for now), our post-patch parse gives us (1583511 + 1750196) / 2312 seconds = 1441.9 DPS, which is ~1.8% less than our previous result of 1466.3 DPS.
So it looks like Sigmund Freud was right: Sometimes a tooltip really is just a tooltip. There doesn't appear to be any real damage buff here. In any case, and as I suspected, Storm Spell doesn't appear to proc from any of the lingering pillar's attacks, but on the upside, Storm Pillar appears to have a higher-than-expected proc rate on the main strike:
Total number of Storm Spell procs: 172
Storm Spell proc rate: 172 / 1000 = 17.2%
That's significantly higher than Magic Missile's proc rate, but of course that little factoid is a little misleading, because Storm Pillar's main strike only occurs once every ~2.3 seconds, whereas MM attacks occur once every ~0.45 seconds. For a more realistic picture, we can include all of Storm Pillar's attacks, because MM and Storm Pillar 'tick" roughly at the same speed if you include the lingering pillar effect:
Total number of Storm Pillar procs: 172
Storm Spell proc rate: 172 / 6000 = 2.86%
Finally, the Storm Spell proc accounts for 151838 / (1735349 + 1750196) = about 4.35% of our parse's total damage. Kind of funny, really, that the Paragon Path's best passive augments the Paragon Path's only At-Will far less than it does every other At-Will.
and also the featured satirical comedic adventure "A Call for Heroes".
Will be (slowly) adding feat (and feature) information as we go. Truth be told, I spent an inordinate amount of time in the last coupla weeks distracted by my Guardian Fighter alt.
Although it's true that we can probably calculate the remaining feat/feature benefits in a straightforward way, I want to be sure that there aren't any irregularities with respect to the stacking of damage bonuses. Cryptic is kinda known for less-than-intuitive stacking rules, in my experience (eight years of City of Heroes, about a year each of Champions and STO, and now a couple months of Neverwinter -- not that any of that lends me any particular credence, mind you.
Just the other day I double-checked HP-bonus stacking and came away with the conclusion that CON modifies base HP, but not anything else. So Toughness (the feat) stacks multiplicatively with CON, which is nice, but HP bonuses from gear only benefit from the feat. (That is, if you equip an item with +480 HP, Toughness will turn that into a ~523 HP boost. CON will do nothing for that item.)
The bit about HP probably isn't new information; I'm sure it's available somewhere else, but it was news to me
What about Action Point gain for using the At-Wills? Putting aside the exploit of zapping at the darkness with Storm Pillar to gain action points, how do each of the At-Wills gain in AP comparatively?
and also the featured satirical comedic adventure "A Call for Heroes".
Ah, good point. I kinda glossed over AP gain because Lamandal and grimah have covered the matter to my satisfaction. The results are in grimah's excellent guide.
But you're right. If this is to be a comprehensive study of all things At-Will related, I should add a note to the OP about AP gains. They're kinda hard to test because the game does a lot of rounding (and the AP bubble mouse-over-display thingie only gives us whole numbers), but perhaps I'll do a little testing later just to see what the results are at extremely low levels of recovery.
Also a good catch!
Using grimah's numbers (quoted below), and if we ignore for the moment the possibility that the game is rounding them up or down, Ray of Frost gives us an AP gain of 9% per ~3.5 seconds, or about 2.5% of a full AP bar per second.
Storm Pillar gives us an AP gain of 4% per ~2.3 seconds, or about 1.74% of a full AP bar per second.
Magic Missile gives us an AP gain of 3% per ~2.25 seconds, or about 1.33% of a full AP bar per second.
Chilling Cloud looks to be tied with Magic Missile.
The differences don't sound like a lot, but over time they'd really add up, which explains everwindgale's experience; Ray of Frost gains AP about 43% faster than its nearest competitor. Of course, standard disclaimers apply: these are all preliminary numbers, and as you can see, they're subject to serious error if our initial values are derived from rounded numbers, but there's a lot of potential ground to be explored here. Even if Ray of Frost only turns out to hold half of the above-described advantage, it'd be a significant thing!
One of these days I need to figure out how the whole log parser stuff works, because I really want to see how my normal lightning enhancement experiment is really working, and what is actualy better to pair with Storm Spell as my buff powers.
Chilling Cloud has 4 components to it: Damage, refresh chill, apply chill, apply damage debuff. Assuming that those 1,212 attacks were 404 full cycles, that would be 808 attacks with the damage and refresh components, and 404 attacks with all 4. (808 * 2) + (404 * 4) = 3,232, and 344/3232 = 0.1064, or about 10.6% proc chance, nearly the advertised proc chance.
Dunno for sure if this is what's causing the numbers to be what they are, but given what I've seen with Icy Rays, it doesn't seem too unlikely. I imagine that Ray of Frost has a similar thing going for it. Damage, refresh, apply, and freeze. Damage, refresh, and apply chill for the first 6 stacks of chill, then all 4 for each hit that attempts to place a 7th stack.
Appreciate the kind words, morrokain. Please do keep in mind, though, that I'm nowhere near finished here; the other At-Wills are catching up to Storm Pillar really quickly. Its damage output is impressive in a vacuum, but there's a very low ceiling on its improvement.
I do think Storm Pillar's gotten a bit of a bad rap, but not entirely without justification: before the patch a few weeks ago, it didn't crit. And although we now know that Storm Pillar's charge time isn't unreasonable in comparison with the activation times on other At-Wills, it does seem rather long when you first try out the power. One of the main reasons I wanted to test At-Wills is that I wanted to nail down the time component; damage is easy to see, but when it comes to measuring small increments of time, the human eye is notoriously inaccurate.
This is an excellent theory! I think you're right. If you don't mind, I'd like to link your post on the front page of the thread.
I guess my dreams of gaining 25% more Ray of Frost procs for investing in Glacial Movement are dashed.
In the case of Ray of Ice, you are completely right in it's ability to generate faster procs than any other skill because of two factors:
- Animation lag is shorter with a constant ray of ice shooting at player vs (windup animation of the player + actual missile attack on the enemy)
- The difference in time between each tick of ray of ice and magic missile gives Ray of Ice an advantage due to lack of animation after initial windup of the skill. This is why Ray of Ice is essentially faster on the actual ticks vs the time it take to complete a full cycle of 6 stacks.
Essentially what I'm trying to say is that Ray of Ice ticks at the same speed as Magic Missile on paper, but in practicality and with consideration of windup + animation, Magic Missile is indeed slower at attacking and stacking procs.
It seems to me that the way critical chance works is more than likely the same for how storm spell works with the difference being that storm spell is activated by ticks + hit and crits by hits only.
Video Review: http://www.twitch.tv/zovya/c/2671017
For what it's worth, my data shows no unexpected behavior with regard to Magic Missile. Both crits and (Storm Spell) procs occur as advertised. I'm not saying you're definitely wrong; it's possible that Magic Missile's second and third attacks don't count for the purpose of crit/proc calculation, but the data suggests that all five attacks count.
Again, the data suggests otherwise. I ran long parses precisely because I wanted to rule out any irregularities in my cast time measurements, due to client/server lag or anything else. On average, Magic Missile attacks significantly faster than Ray of Frost (~2.2 attacks per second versus ~1.7).
On the other hand, Ray of Frost does proc (Storm Spell, but also to a lesser extent weapon-enchant effects, as we'll see in my next update) way more often than Magic Missile. So far, the most likely explanation I've seen for that phenomenon came from itsamepikmario, earlier in this thread: basically, he asserts that Ray of Frost (and Chilling Cloud) proc more often because the game counts every opportunity to apply/refresh chill and any other buff/debuff as a proc opportunity.
I have to do more testing to put that theory through its paces, but it seems to fit the data. I can say right now that although Ray of Frost procs weapon enchants more often than you'd expect, based on bona-fide attacks logged, Ray of Frost's weapon-enchant-proc advantage isn't nearly as high as its Storm Spell advantage. So I have to guess, tentatively, that there are some quirks unique to Storm Spell (or perhaps unique to random-event Class Features).
All of that said, I do appreciate your feedback. Neverwinter has more than its fair share of mechanical oddities, the explanation for which players are, understandably, prone to invent any number of interesting theories. That's why this thread exists.