Stats Revamp 1.7 - Light

Discussion in 'Stats Revamp Archive' started by spord, Jun 9, 2017.

Thread Status:
Not open for further replies.
  1. Green Lantern Fadi Loyal Player

    yes this!!!!^^^
  2. Swiftduck Loyal Player

    I'm already on the fence from completely leaving this game. This iteration is worse than live. In live is a great balanced power, what the heck is this?
    • Like x 2
  3. xD25x Dedicated Player


    It was in a nice place and then the devs made unnecessary changes. 1 step forward 2 steps back.
    • Like x 4
  4. VioletSorceress Committed Player

    Seems like Devs tried to incorporate same set of regulations that got weapons all messed up to combo powers as well.
    • Like x 3
  5. Mighty Committed Player

    Right? Just completely scrap the adjustment to combo powers, it makes zero sense, was not wanted, needed, or asked for. The iteration of light on test right before this was very good. It was power hungry at the high end, but you had options to throttle down the power consumption for DPS that was on-par with what most other powers seem to be at. Hell all of the feedback up to this point was "little power thirsty, but everything else is good!" And yet they had to step in and fix something that was fine.

    One step forward, four steps back.
    • Like x 5
  6. CrappyHeals Devoted Player

    This last update to combos has really took a dump on some combo powers like light and atomic. I honestly think we should not be using this formula for combos and should roll it back to the prior version. Combos all can't be balanced under the same rules cause they all don't work the same. Things were just fine before this change, maybe somethings needed slight damage adjustments but thats it. I know you guys are trying stuff and thats cool but this is just one of the things that you have to say "well that didn't go as planned"

    We really need to stop with this blanket formula stuff and balance things separately cause what works for one power or thing might not work for another as we keep seeing time and time again.
    • Like x 4
  7. Mighty Committed Player

    Hey Spord, caught a couple things while giving light a more thorough lookover.

    Bug 1: the first hit of construct chainsaw out of hand clap, and out of impact does not register damage.

    Bug 2: Power ram appears to have splitting issues. It does non-crit 11.5k mean damage on one target, 7.5k per target on three targets. I was under the impression AOE had been bumped to split after three targets.

    Bug 3: I'm getting a similar weird split on construct whip thrash where a third of the damage is bracketed in the 2.5k range, and the rest of the damage is bracketed in the 1k range, again non-crit mean damage with no data points inbetween.
  8. light FX Steadfast Player

    Exactly right. I know many players had concerns about the combos even a few builds back. I posted about having concerns with combos awhile ago. I saw a potential problem and what do ya know it happened. This 1 step forward 2 steps back stuff is becoming very frustrating. Whenever an ability or power needs a slight adjustment it always ends up being way overdone. 1 extreme to the next :confused: and thats been happening for a long time now. I have a theory why it happens but i cant really post it w/out getting into some hot water :p
    • Like x 1
  9. Mighty Committed Player

    Hey Spord,

    I’ve spent some more time running numbers this weekend on both light, and on other powersets to try and get a feel for how to fix some of the issues we’re seeing. This’ll be a long post, but hopefully worthwhile and constructive, backed by numbers. I have saved the parses used in my data crunch, and if you need them, they’re yours.

    First I want to put down a few thoughts:
    • I love this game, and this is by far my favorite power. I want to see the revamp do right by hard light, and I want to see the revamp do right by the game and help ensure its longevity and success. If there’s any question about my motivation, it all comes back to this point, to make this revamp successful for light and for DCUO.
    • In the past I’ve done weekends-long analysis like this and it turned out to be a waste of time for me because of communication issues. I’m putting any past slights or frustrations aside to do what I can to help you make this revamp successful. I’m willing to do whatever, even jump into a voice chat if needs be, even to keep that discussion private if you prefer to discuss how we can improve light.
    • “Improving light” is a vague and subjective term to be sure, so I’ll clarify what I mean here. Light on live right now struggles from an enjoyment factor. To be candid, it’s not that fun or challenging to play. There’s virtually no skill ceiling, and that’s a shame given how the power played prior to Game Update 36 and the changes to clipping, jump canceling and the intro of weapon mastery. Light on live also has some damage parity issues because of the melee problem I know you’re cognizant of. To keep pace with other “melee” damage rotations you have to be on top of the boss and there’s more risk in that playstyle than other powers’ melee. Light on test as it is now suffers from both a damage parity issue – it just doesn’t keep pace with other powers – and a power consumption issue – trying to keep pace burns way more mana than other powers.
    SO, when I talk about “improving light” I’m talking about bringing back a fast-paced, clip-happy playstyle reminiscent of pre-GU36. I’m also talking about remedying the damage parity and power consumption parity issues. To the playstyle and pace particularly, I feel most veterans would agree that the golden days of light were back before weapon mastery and the extended combo AM, when light played fast and creative, and had a high, difficult to master playstyle that rewarded players for practicing their clip speed and timing, and rewarded players for the higher than average power consumption associated with it. I know we can’t completely go back to those days, but the previous iterations on test were close enough to bring me back to the game and get me hooked, and I think we can get it close enough to maintain parity and warm the hearts of salty old hard light vets like myself.

    I think I understand the constraints you’re trying to operate under, and so I’m trying to work within those as well. It seems like, the unique problem with light is depth and breadth of combo options, the ability to clip combo animations almost instantly, and the relatively short animation times of tray powers. I can see how it’s difficult to keep power and damage parity, and the fast-paced clipping style, and keep the creative uniqueness of the power intact. I feel that was part of the consideration in the nerfs to certain light (and other combo) powers that went to test last week. I’ll venture to work within that reality.

    That said, here’s my work. I put on 175 vendor gear to reach CR 201. I’m modded straight might in my gear sockets with the exception of might and power special forces mods in my two yellow sockets. I’m modded full might special forces in my generator, as well as full SF power mods. I’m capped out in might/power and all crits in my stat points. I’m wearing a max damage hand mod, core strength chest mod, and escalating might neck mod with replenishing procs in my weapon for more power. I tested all loadouts with two trolls spamming power.

    I began testing first by looking at the mean damage for AOE powers with a PI running (Ignore single target for now because I’m still mining the date, it’s slower because of the snap and grasping hand dots and the way the parser reads them). Then I went back and tested constructs with a PI running. I charted the mean damage from the DCUO log analyzer, and then divided the mean by the power cost to create a damage-per-power cost number for comparison’s sake.
    Here are my results for powers from the tray:

    [IMG]

    Here’s my results for constructs:

    [IMG]

    Given more time, I’d like to go back and chart animation times and factor those in, but I’m not sure it’ll be entirely necessary, because I know how many times I can fit a particular combo into a 12-second PI duration. More on that later.

    With this information I wanted to compare light to some other powersets to get a feel for the parity in terms of both damage and power consumption. While doing this, I settled on rage as the best model for a few reasons. First, rage and light are combo power cousins with similar concepts (some differences, absolutely, but more similarities exist here than with other powers). Both have ranged and pure, lunging melee combos. Both are traditionally thought of as faster-paced, clip-friendly powers. Second, after playing around with rage this weekend, I think it’s in a pretty good place overall. The power consumption isn’t too severe, there are multiple options for loadouts and rotations, and the damage at range and in melee seem to be in the right place. If really pressed, I think the ranged damage could be brought up a tad, but that’s it.
    • My target bracket for DPS is 33-38K DPS ranged in this gear with these stats. This discounts supers and trinkets, etc. It’s just a basic “what’s the rotation do?”
    • My target melee DPS bracket is 40-45k DPS. I base these thresholds off of my own parsing, and conversations with peers about what they’re achieving with other powers in the same spec.
    With that in mind, here’s what I tested on rage’s main ranged and melee AOE abilities, focusing specifically on dreadful blast and outrage and their corresponding combos.

    [IMG]

    [IMG]

    As we can see, at both range and melee, rage’s staples get much more damage out of their power consumption than light. Bombardment is a bit of an odd duck in the chart because the parser doesn’t identify the tray bombardment from the construct one. That said, I parsed 33-36k DPS with a Dreadful blast-Bombardment-Plasma Wretch ranged rotation, and 40-45k DPS with a Galling eruption-Outrage-Dreadful blast melee setup. At no time was I sweating for power the way I have with light.

    Going further, I thought it important to not just look at individual moves in a vacuum, because they’re part of a combo system, I wanted to compare the combos as a whole. I think just making sweeping changes on powers in a vacuum leads to the type of disparity we see in light on test currently. I started with melee because I think the animation times between light melee combos and outrage are very similar, thus making for easier comparisons. Here’s my chart:

    [IMG]

    As you can see, no light melee combo comes close to keeping pace in either damage output, or damage to power efficiency, and all combos save for ram-whip are essentially similar in length and all involve lunging the target and being immediately at risk.
    Here, I started playing with damage and power and came up with three proposals, In all cases, I set aside the damage and power cost ratio of individual constructs and powers and I look at the damage-to-power of the combos as a whole:
    1. Change power claw to hit in a spherical AOE when played from the tray. It’s currently single target. Just doing this would allow players to cast claw, combo to final claw, and finally combo into entrap-swing. If power claw’s single-target damage was extended to three targets (assuming power would probably be bumped from 200 to 300) it would create an outrage-outrage-final measure-like combo that did comparable damage at a comparable power cost. Claw could even get the same cooldown as outrage to ensure it’s not being clipped with itself, rendering the combo OP.
    2. Greatly buff power chainsaw and the second melee tap of the chainsaw combo so that it totals approximately 75K damage at 500 power over the three taps. Power chainsaw is currently hitting for about 7,500 per target which is pretty extremely sad for a 300-cost melee power. The follow-up tap does decent damage, but the third tap does a mere 3,000 damage approx. For such a cool animation and power, it’s the worst light melee combo in terms of power efficiency, and the total damage is pretty weak as well. Again, chain could get a cooldown similar to outrage, forcing players to clip it with something other than itself.
    3. Entrap’s power cost should be reduced to 200 from 300, as it’s very weak in terms of damage per power cost, even for a ranged attack.
    Enacting these three changes, the combos would look like this in relation to outrage:

    [IMG]

    The only real parity issue I see with this is giving light two outrage-like moves to pair together, but at the same time running both combos would require separate armories for melee and ranged DPS, as well as light does not have the same damage mitigation potential as rage, nor does it have a berserk-like super charge. More risk, slightly more reward.

    • Like x 10
  10. Mighty Committed Player

    ctd.

    Moving on to range I ran into an issue modeling off of rage in that the dreadful blast combo is a little longer than your standard light power-construct/power-construct setup. I didn’t want to do a hard-and-fast comparison for fear of changes to light leading it away from the fast pace playstyle so many of us want to see it return to. So I used the damage-to-power-cost ratio from dreadful blast as a guideline, and instead went about reverse-engineering what was the best light combo on test prior to the recent fix. See I think it’s not just enough to reduce the power cost, because the damage will still underwhelm, moreover it’s not enough to simply buff the damage, because light uses too much power relative to other powersets.

    [IMG]

    The best combo I found on the last iteration was a simple Entrap-impact/Ram-Fan and then repeating R-F nine more times before the 12-second dazed PI wore off. That setup, while extremely power-hungry, would net me between 38-43K DPS consistently. I did feel a little disappointed that the best light ranged combo involved spamming one simple combo to death (arctic gust anyone?). So what I did is I figured I could slip 10 combos into a PI setup, and then I looked to see if there was a way to responsibly buff certain powers and constructs within that chain, and bring down their power costs, and have it fall in line with other ranged DPS parses.


    I found that by cutting construct power costs on impact and fan from 100 to 50, and then buffing their mean DPS by 75%, from approx. 2600 per-target damage to 4500 per-target (here I’m lumping the three impact tics into one for organization’s sake), we can goose light’s ranged damage up to approx. 36k DPS to put it in line with other powers’ ranged AOE capability, and also easing off the power consumption slightly.


    [IMG]

    I think these fixes could help improve light by making it a competitive, reward and fun power to play, and a staple of a successful stats revamp. I’d love to get the opportunity to play those changes and see how they feel damage- and consumption-wise, and would be glad to help offer input on future tweaks until we get it right. I think it definitely will take stepping away from a hard-and-fast power-per-damage-per animation time approach, because I don’t think the blanket changes will really address the issues light has. I’d love to hear your thoughts on this, and thanks for reading.


    TL/DR
    1. Change power claw to hit in a spherical AOE when played from the tray. It’s currently single target. Just doing this would allow players to cast claw, combo to final claw, and finally combo into entrap-swing. If power claw’s single-target damage was extended to three targets (assuming power would probably be bumped from 200 to 300) it would create an outrage-outrage-final measure-like combo that did comparable damage at a comparable power cost. Claw could even get the same cooldown as outrage to ensure it’s not being clipped with itself, rendering the combo OP.

    2. Greatly buff power chainsaw and the second melee tap of the chainsaw combo so that it totals approximately 75K damage at 500 power over the three taps. Power chainsaw is currently hitting for about 7,500 per target which is pretty extremely sad for a 300-cost melee power. The follow-up tap does decent damage, but the third tap does a mere 3,000 damage approx. For such a cool animation and power, it’s the worst light melee combo in terms of power efficiency, and the total damage is pretty weak as well. Again, chain could get a cooldown similar to outrage, forcing players to clip it with something other than itself.

    3. Entrap’s power cost should be reduced to 200 from 300, as it’s very weak in terms of damage per power cost, even for a ranged attack.

    4. Cut construct power costs on impact and fan from 100 to 50, and then buff their DPS by 75%
    • Like x 9
  11. Amontron Well-Known Player

    hey mighty. can you give your loadout for me to test please? mine hits 20ks, im pretty sure melee does alot more damage than the range combo. i tried few times.
  12. Mighty Committed Player

    Sure thing, when I parsed that I was running chompers for the PI and a ram-fan spam with a rifle. Looked like this:

    Rifle nade/chompers-nade/Ram-Fan/Ram-Fan/~

    Was able to pop about 10 R-Fs until I needed to reset my PI.

    You may not be able to replicate that currently because on Friday construct fan got nerfed into the ground. It currently hits almost as hard as a rifle tap at 100 power cost. Also, I'm pretty sure I was CR 186 when I ran that because I was testing with the gear box they give you on test. Current testing meta is to go full special forces and be 201 CR with special forces generator mods.
    • Like x 1
  13. Mighty Committed Player

    Bug: when running with a mental DPS in the team, if mental guy pops his PI conversion power (telekinesis for example) to convert dazed PI to terror PI, it converts everyone's dazed PI into terror, completely wiping the effect from the target. We tested this by popping chompers and then immediately casting pyro on the same targets. The visual dazed effect dropped off, and the damage for powers taking advantage of dazed dropped to what it would be had there never been a dazed PI on the target at all.

    I posted this in the mental thread, Spord, but I wanted to throw it in here, too to get more eyes on it. Plus you're the man, so I know you can get it fixed.
    • Like x 2
  14. Amontron Well-Known Player


    thank you mighty i will run this loadout on test and i will tell you what numbers i will have and what changings that might happen. i will test with the new 201 gear and mods on 3 sparrings targets. so, i suppose you are with Hybrid mastery right? did you tested HL with the Power Mastery? im using it and my regen is better since i use alot more of the power constructs and less or no weapon at all.
  15. Mighty Committed Player

    I was power mastery. I haven't played much with hybrid on anything because it's not really my preference. I still sneak in rifle taps where I can, though.
    • Like x 1
  16. Maxwill Committed Player

    I read some of your long post, and I'll reply to this that I don't like.Entrap powercost should remain a 300 and have the damage increased if it's really weak in terms of damage per power cost.
  17. Mepps Sr. Community Manager

    All posts in this thread have been reviewed by the team.
    • Like x 1
  18. spord Developer

    1) Can you elaborate on if you were already at melee, flying, or at ranged? I cannot repro this but maybe there is a specific way you were getting this to happen.
    2) Splitting is working correctly. 11.5 * 2 = 23 / 3 = 7.6666. You get full damage (11.5) each on two targets (23). Once that third target came in it will split that 2x full damage (23) by the number of targets.
    3) Splitting is working correctly here too
  19. spord Developer

    I'm really glad you love the game and enjoy Light powers. You clearly put in some work over the weekend. There are a couple of factors to keep in mind for these types of comparisons. Abilities with crowd control will do less damage than those without it. Chainsaw & Entrap both have crowd control. Animation time plays a large factor in the damage calculations. As you said in another post, combo animations don't matter as much because they are being clipped. That is why these changes were made. Because they can be clipped heavily we can't give them a damage multiplier as if they were longer. If we do then we end up with combo set powers dealing 30% more damage than others. That being said we obviously don't remove the animation time calculation completely either.

    So the feedback we could really use more of right now is full on best of rotations vs other power sets best of rotations. This will ensure that no one set is completely above the rest. Plus how many rotations in a powerset can be in the best of category.

    As for your comparisons to Outrage. The abilities that you want changed would need to no longer be single target, have no crowd control, and cost more power. That would make them exactly like Outrage. Is that what everyone wants? Maybe just Chainsaw because Entrap without the encase seems weird.

    Also Fan was doing double damage to enemies that were Dazed. Clearly that was busted and fixed when we made the combo time adjustments. Sorry I forgot to add a patch note I was doing a lot of adjustments at once.
    • Like x 3
  20. Mighty Committed Player

    Right on, thanks for the quick feedback. I'm on the fence with entrap, because traditionally encasements are kind of obnoxious. If it didn't encase I honestly wouldn't have a problem with that, but on this particular issue I feel uncomfortable speaking for all the light folks. Entrap really would be odd if it didn't encase and you couldn't swat ads around the room with the bat. Maybe it's obnoxious for serious content CC, but it's fun. I don't want to be the guy who ruined that. A change to at least chainsaw would be nice because we have these handful of melee combos but we're not getting melee bonus (or whatever the preferred nomenclature is there) damage off a lot of those abilities/constructs. I'm curious on power claw there, what's your thoughts on keeping it single target? Right now light has a pretty good arsenal of single target (dare I say we have the best single target ability collection). I'm just thinking that the primary aoe melee combo rotation for light could be either: claw-final claw-swing/chainsaw-chainsaw-chainsaw, or sub in entrap-swing in there somewhere.

    I'll work to get you best-of parses, but I do have questions there, because some of us have noticed that some melee weapon combos (MA smokebomb) do pretty incredible damage, and it seems like weapons are a bit in flux right now. For the time being, would you want straight pftt+rifle taps? Also is it easier if I just upload video with my parser running?

    Final thought for now, right now lightweight (the original, not the SC generator), has some really good potential, but the lone issue is pretty big in that it moves anything it hits that's not bolted to the floor, including tanks. I'm guessing because it's technically an object is why this happens, is there any way around this? It'd be great and useful in a lot of spots but for it knocks the tanks around with it.

    Thanks again, I know there's a ton of variables y'all are working with, the complexity and the importance of the revamp are not lost on us. You have our appreciation.
Thread Status:
Not open for further replies.