User talk:Devuu/Elementalist Damage Project

I don't use tb, but I doubt it's smart enough to take deep wound into account. Every time a mob dies with a DW you cause, it's essentially 100 uncounted damage. -- DANDY ^_^ -- 00:12, 11 May 2020 (UTC)
 * Thank you, and that is a great point. I don’t think it accounts for DW, which would definitely account for the lower “listed” damage of AP, and explain why the total damage done for all characters is lower in the AP setup. Based on this, AP is probably closer to competitive for raw damage to EA, but it’s advantage is still clearly in utility.— Devuu ((Yuko Asakura [PhD]))  00:33, 11 May 2020 (UTC)

Testing Areas
I'm looking forward to see more test results. Although ravenheart gloom will, of course, give different results than a lot of other areas. It has unusually strong anti-melee (not just for melee attacks, also for point-blank casters), has a larger foe density than the majority of content and larger armour ratings due to the large base level of foes. But there may be some insights to gain from the place nonetheless. Happy testing! --Krschkr (talk) 01:18, 11 May 2020 (UTC)
 * Ravenheart was chosen specifically for these reasons:(1) it demonstrates the AoE capabilities of a build very well—-if you can’t clear cave fast enough your ST will be overwhelmed and you will wipe, since there is no real tank; (2) has high enough AL levels that you can see the impact of elemental v. armor-pen (I.e., for ele to outperform Dom Mesmer hero, the build must actually be capable) but does not favor one element too high over another (for example Titan quests or certain EoTN dungeons, etc.); (3) does not punish either Enchants or Hexes overly hard, so one can’t argue one build is given unfair advantage; (4) not overly punishing to minor human error, but still difficult (was originally going to do Foundry HM, but it is simply too easy to wipe on double pulls or due to user error, and I am not immune to making mistakes, especially for builds or play-styles I am not used to); (5) does not require overly niche builds to complete successfully, like certain Dungeons for example. When selecting I did not choose an more normal or less challenging area to account  for General PvE because I don’t think there is much that pressures a build to perform quite like as a HM Elite area. I may consider smaller areas or something like certain HM missions to test speed as well as damage. I am interested in any suggestions people might have, especially for when I eventually test out the E/A dagger ele.—Devuu  ((Yuko Asakura [PhD]))  06:52, 11 May 2020 (UTC)

Toolbox
The damage measuring is inaccurate. On top of general accuracy issues and not taking account of deep wounds (as Dandybot pointed out) there are some more issues to keep in mind. A bit more on the last point. Degeneration is entirely ignored by the damage meter. This causes an increasing discrepancy between the actual damage inflicted and the reported damage. Example 1:
 * It really is pure damage and does not tell very much about the qualities of a build. Support for other builds, utility, shutdown, healing and so on don't appear here.
 * It counts overkill damage.
 * It ignores degeneration.
 * Scenario: You fight a group of 6 foes standing in the area of the targeted foe, leading with cry of pain. One foe dies after 2 seconds, two after 6 seconds and 1 after 8 seconds.
 * Reported damage (for this skill): 300 (6*50)
 * Actual damage (for this skill): 720 (6*50 + 2*10 + 2*6*10 + 8*10 + 2*10*10)
 * Inaccuracy: 420 damage (−58,33%)
 * Inaccuracy: 420 damage (−58,33%)

Example 2: If your build inflicts degeneration (most likely burning) the reported damage will be off. Due to toolbox's failure to detect degeneration you have to guess how much damage slipped through the detection if you want to compare the pure damage output. And in case of burning that's up to 14dps/foe, depending on the used skill on adjacent, near or foes in the area. There can be quite a lot you're missing. --Krschkr (talk) 01:36, 11 May 2020 (UTC)
 * Scenario: You fight a group of 8 foes with 100 armour and use Searing Flames on it. All foes die simultaneously after 16 seconds. You suffer from [[Migraine], so you'll get off searing flames four times: After 4, 8, 12 and 16 seconds. The first and third cast only inflict burning, the second and fourth only deal damage.
 * Reported damage: 848 (8*2*53)
 * Actual damage: Roughly 2088 (8*2*53 + 8*7*14 + 8*4*14 + 8*1)
 * Inaccuracy: 1240 (−59,39%)
 * Inaccuracy: 1240 (−59,39%)
 * Great feedback. I think this portends promising results for both builds relative to Dom Mesmer heroes and to a small extent evens out discrepancies between the ones tested so far. This might make air seem slightly better than fire on the damage tracker, which is a consideration I will keep in mind in tracking results.—-02:20, 11 May 2020 (UTC)Devuu ((Yuko Asakura [PhD]))  02:20, 11 May 2020 (UTC)
 * Just wanted to say that toolbox is accurate for pure damage. It can't measure degen, and it can't measure the damage from deep wounds. I've done some testing for hero bars, which showed me a big diffrence between Air/EA Fire, and Searing Flames. A thing to keep in mind is that fights aren't going to last long with a good hero team, so degen usually doesn't really have time to shine. If you want to use toolbox for testing the various player ele builds, you should also look at the clear time. You can't really measure the impact of KD+Deep Wound, and enemies might die quicker, which means more chance of succes and less spells that get "wasted" on low HP targets. ZStepmother (talk) 08:32, 11 May 2020 (UTC)
 * The thing is that I personally don't see the merit of toolbox at all when testing hero team builds. Task completion times and stability aren't reported by a damage tool. Even if a player build does 30% of a team's damage it can slow down the team compared to a player build that does 2% of a team's damage, even without changing the heroes. Because the 2% build might be a heroic refrain paragon who boosts the entire rest of the team by a lot and stabilizes it far enough to play very aggressively. The 30% build, on the other hand, can be a minion master who needs to make minions after combat and prepare before combat, resulting in a lot of wasted time. --Krschkr (talk) 10:54, 11 May 2020 (UTC)
 * Everything you said is valid, but Toolbox damage meter is still a good way to measure pure damage output. Most ele builds are just that: pure damage. For the tests here, toolbox damage meter is a pretty valid tool to measure the effectiveness of a build. It's not right to focus on toolbox alone, you always have to look at the bigger picture. Like I said, the AP build does offer something extra: shutdown (Single target: YMLAD, EVAS and simply killing single enemies | AoE: meteor shower), which is why I suggested also looking at the clear time.
 * More specific for this build: Meteor shower really shines at the cave (and maybe at darkness), but outside of that it's not really useful. I definitely do think that shutdown is really good in doa, but I don't think its worth gimping your entire build for it. Just swap 1 Esurge for Panic, and take an actual effective damage bar as the player (so not AP). &mdash;The preceding unsigned comment was added by ZStepmother (talk &bull; contribs) 11:55, 11 May 2020‎ (UTC).
 * Thanks for your feedback. You're right that Meteor Shower shines in most of DoA because of how powerful lockdown is against the powerful opponents. However, I disagree with the idea that running MS gimps a player build. If you are going to run Fire with AP, Meteor Shower is a must.  Running a standard Nuker with AP (i.e., GOLE, Rodgort, LF or Fireball, etc.) is basically a gimped version of Ele Attunement that will never win in raw damage and banks heavy on EVAS/FH/YMLAD utility.  Further, as evidenced by the raw numbers, there is nothing showing that the AP MS Spammer cannot pull its weight, especially given that deep wound is not calculated in the damage tracker.  What's more, if we want to move the discussion to speed or overall effectiveness, the AP setup was really smooth and led to faster cleartimes (look at how much less damage was done across all characters, as this indicates how long the runs took). Finally, to address one build getting greater benefit in Ravenheart (i.e., that MS is inflated due to Cave value, etc.) it's hard to argue that running Ele Attunement (with spammable Area and Nearby damage) doesn't get a huge benefit in Ravenheart, especially considering that I brought AOS which is probably the best skill possible for the Cave (other than EBSOW, arguably).  If anything, I think ths testing shows that AP is NOT a gimped or trash build for difficult content, but can be really effective and rewarding.  It just brings to the table different things than a standard Elementalist nuker does. I am also looking forward to testing out AP Earth, Invoke Lightning, Dagger Spammer and possible some other now archived builds. Thanks again for the feedback--I'm really hoping we can gain some insights here.  -- Devuu  ((Yuko Asakura [PhD]))  17:59, 11 May 2020 (UTC)
 * The total amount of dealt damage doesn't say much about the actual clearing time though. It means that you fought more foes or had more overkill damage or had a lower share of unidentified damage or that you were better at suppressing healing. The latter incidentally is the strength of AP builds. --Krschkr (talk) 18:07, 11 May 2020 (UTC)
 * For timing, I will say that I believe that the cleartimes were slightly faster for AP. It is hard to be precise, because for a number of these runs, I had to go afk in the middle of some of these, so /age is not a great indicator. The fastest clear-time I had was ~20 minutes, with the 48.5k dmg AP clear (uninterrupted). Unfortunately, with real life and pandemic situation, things are somewhat unpredictable. One thing I will try to keep in mind, in addition to numbers, is over all performance and trying to keep track of different strengths that might not be reflected in all builds. Maybe I can come up with a more subject metric/measurement system too.  I will keep you informed on that development. -- Devuu  ((Yuko Asakura [PhD]))  19:25, 11 May 2020 (UTC)
 * I'm not saying that Meteor Shower gimps your bar. I'm just saying that I think Meteor Shower is the only reason you get good results here. HM DoA without any shutdown (like Panic) is not advisable, and you don't have any shutdown if you don't run that AP bar. The utility value in that situation is real, but can easily be solved by swapping an ESurge for a Panic, and bringing a player build that actually deals damage. I think DoA will just inflate the utility that AP bar brings, because in general PvE, your fights don't last long enough for Meteor shower to be effective. I'm not saying AP isnt an effective build, I'm just saying that there are much better options (from my experience running all these bars). ZStepmother (talk) 20:16, 11 May 2020 (UTC)