GU50, Tunso Testing, and Player Testing

Discussion in 'War Room (Powers, Artifacts, & Builds)' started by Sore, Jul 16, 2015.

Thread Status:
Not open for further replies.
  1. Remander Steadfast Player

    See my other post.
  2. CrappyHeals Devoted Player

    But all we needed was one example of how the tunso method was wrong and we could have been on our way to getting stuff done.

    Edit: and seriously how could he think the results from a test like that would be right, might power vs might power yea but not comparing others.
  3. Remander Steadfast Player

    It's not wrong because of the parse sizes, though.
  4. Sore Steadfast Player

    Since many rotations are 24s, you should be doing the math anyway. I think we are going to be done with the era of eyeballing parser results. Yes it gives you a tighter range to discuss balance but the overall powerset comparison between powers is more easily analyzed with a single hard number like a mean. It lends to all kinds of interesting charts that range-based discussions don't cover well. Frankly, people get too much sticker shock from the big numbers anyway. Just because you technically can get 35K with something in a 30s sample doesn't mean a lot because you're not getting it often. And honestly, you're probably only getting it because the 30s sample has the burstiest portion of two different 24s rotations all in one sample.
    • Like x 4
  5. CrappyHeals Devoted Player

    No its not but its easier imo to get consistent data using 30s imo
  6. CrappyHeals Devoted Player

    Oh i understand that completely, powers work way dif in game. With celest i can hit 41k on AoE so on paper i should poop all over everyone, but in game its a totally dif story. Thats one of the reasons i don' t like what they did to ice. its to easy and fast to be putting out those numbers

    Edit: and you always have to take AoE parses with a grain of salt but single you can take more seriously.
  7. CrappyHeals Devoted Player

    Step away from all the graphs and charts guys they are not really needed, this is to simple to be doing all that lol:p
  8. Sore Steadfast Player

    But fundamentally these tests are pretty easy to execute. We are one test away from full coverage for the Tunso Tests and the completionists in us just want to see it through. We are pretty sure the "Tunso Test" will not be the standard methodology in the future. The Live Test is methodology is pretty decent in my opinion. It would be better if PC Test offered Prec/Might VII mods. Once we've checked our boxes on the Tunso Tests, we will likely move our focus to the Live Tests. And I don't mind because the Tunso Tests only took like 5 days to capture which in my book is a blink of the eye. I assume the Live Tests will be pretty quick too.

    But again, it all is getting ahead of ourselves and we know it. The first thing we need to know is how Midrange will be addressed. The community was very loud on insisting it exists and demonstrating why/how it exists.
    • Like x 4
  9. not_again Dedicated Player

    Are these test going anywhere? I see sore and radium in this thread but are you guys able to give feedback and receive feedback from tunso? Is he receptive to the proposed mid range test? Just want to make sure all this work is not going to waste like the testing for GU 50
    • Like x 1
  10. Sage-Rapha Steadfast Player

    Sorcery Ranged

    Circle of Destruction, Shard of Life, Offering, Fury, Transmutation, Soul Well.

    Things to note:
    I didn't include Weapon of Destiny since the damage difference is minimal at best.
    To really Gauge Fury's potential I had someone apply Bad Karma with Shard of Life on the 8 targets.
    There were times Fury was low on power so I had marked where he was low.

    Weapon:
    Handblasters
    preferred combo: Scissors Kick

    Sequence:
    Spawn Fury, Lay Circle and Soul Well> Scissors Kick clipped by Shard and Transmutation Interchangably until Offering is off cooldown.

    Parser Summaries from Combat Log
    [Summary] Damage [10.0s] 19386/s - 193856 total - 188 hits (8085 max) - 27 (22.0%) crits - 8 targets (Fury unloaded just right)
    [Summary] Damage [10.0s] 18304/s - 180421 total - 173 hits (9260 max) - 48 (32.0%) crits - 8 targets
    [Summary] Damage [10.0s] 16352/s - 163524 total - 135 hits (8216 max) - 37 (27.4%) crits - 8 targets
    [Summary] Damage [10.0s] 14221/s - 142212 total - 177 hits (7040 max) - 49 (27.7%) crits - 8 targets
    [Summary] Damage [10.0s] 15714/s - 157136 total - 140 hits (5894 max) - 45 (23.6%) crits - 8 targets
    [Summary] Damage [10.0s] 11952/s - 119515 total - 150 hits (7270 max) - 48 (32.0%) crits - 8 targets (Fury low on power)
    [Summary] Damage [10.0s] 14076/s - 140760 total - 127 hits (8981 max) - 46 (26.1%) crits - 8 targets
    [Summary] Damage [10.0s] 17140/s - 171402 total - 151 hits (7775 max) - 42 (29.0%) crits - 8 targets
    [Summary] Damage [10.0s] 15411/s - 154108 total - 146 hits (6793 max) - 20 (28.1%) crits - 8 targets (Fury Hoarded power then let loose)
    [Summary] Damage [10.0s] 16190/s - 161904 total - 130 hits (8894 max) - 44 (25.4%) crits - 8 targets
    [Summary] Damage [10.0s] 18417/s - 184170 total - 137 hits (8509 max) - 37 (27.0%) crits - 8 targets
    [Summary] Damage [10.0s] 19005/s - 190049 total - 154 hits (9645 max) - 36 (23.4%) crits - 8 targets
    [Summary] Damage [10.0s] 20965/s - 209654 total - 157 hits (9020 max) - 37 (23.6%) crits - 8 targets (Lucky Crits)
    [Summary] Damage [10.0s] 17759/s - 177586 total - 136 hits (8672 max) - 41 (30.1%) crits - 8 targets
    [Summary] Damage [10.0s] 20972/s - 209724 total - 188 hits (9828 max) - 46 (25.0%) crits - 8 targets (Fury was on fire here)
    [Summary] Damage [10.0s] 19364/s - 193640 total - 144 hits (8121 max) - 40 (31.1%) crits - 8 targets
    [Summary] Damage [10.0s] 18692/s - 186920 total - 150 hits (7278 max) - 42 (28.0%) crits - 8 targets
    [Summary] Damage [10.0s] 16353/s - 163532 total - 159 hits (9396 max) - 44 (27.7%) crits - 8 targets
    [Summary] Damage [10.0s] 16139/s - 161389 total - 117 hits (7792 max) - 31 (26.5%) crits - 8 targets (Lucky crits)
    [Summary] Damage [10.0s] 14063/s - 140627 total - 152 hits (4292 max) - 34 (18.0%) crits - 8 targets
    [Summary] Damage [10.0s] 15731/s - 157305 total - 162 hits (6958 max) - 47 (29.0%) crits - 8 targets
    [Summary] Damage [10.0s] 18159/s - 181594 total - 137 hits (8204 max) - 34 (24.8%) crits - 8 targets
    [Summary] Damage [10.0s] 16045/s - 160449 total - 162 hits (7771 max) - 37 (22.8%) crits - 8 targets
    [Summary] Damage [10.0s] 12130/s - 127118 total - 131 hits (5709 max) - 36 (23.8%) crits - 8 targets
    [Summary] Damage [10.0s] 15103/s - 151027 total - 140 hits (6985 max) - 39 (27.9%) crits - 8 targets
    [Summary] Damage [10.0s] 12704/s - 127039 total - 179 hits (7523 max) - 53 (29.6%) crits - 8 targets (Low power Fury)
    [Summary] Damage [10.0s] 15047/s - 199469 total - 118 hits (8752 max) - 27 (22.9%) crits - 8 targets
    [Summary] Damage [10.0s] 16245/s - 112451 total - 148 hits (3199 max) - 41 (27.7%) crits - 8 targets
    [Summary] Damage [10.0s] 17578/s - 174785 total - 190 hits (7404 max) - 27 (22.8%) crits - 8 targets
    [Summary] Damage [10.0s] 14878/s - 148777 total - 137 hits (4803 max) - 39 (28.5%) crits - 8 targets
    • Like x 1
  11. Rokyn Dedicated Player

    Hey Sore and Radium, I can't get to Earth AM mid range Tunso Test right now since I'm kinda bloated with work. I will see what I can do tomorrow. The Live tests may have to wait till Thursday or something since these take some time to get right as you know.
  12. Sore Steadfast Player


    It's okay. Thanks to everyone involved we have just covered Melee and Range using Tunso Tests for all the powersets. We have minimal (insufficient) coverage on midrange and I think that's acceptable given we're awaiting a final decision from Daybreak. I anxiously await how they'll handle it just like everyone else.

    There's a lot of good going on with this thread. We've proven a lot. We've proven that midrange needs attention and we've got that message through to the devs (trust me). We've proven that testing methodology needs its bar raised. That includes from what we've done before. That includes what we're doing now.

    We've demonstrated as a community that we can execute these tests methodically. We hit every powerset across two different distances in 5 days. We've built energy and excitement around this effort. We've built momentum and I know we can carry that forward into the next round.

    We will have something to show for this (trust me). I have some i's to dot and some t's to cross before I can unveil the results of this effort. It won't be the panacea for where we are today. What it will also show is what we can do with our future efforts as well. In addition, I can guarantee the devs are listening to what's going on right now on multiple fronts (midrange, methodology, and balance). Not everything will unveil in a short number of days. Not everything will be fully addressed in a short number of weeks. Think of it like peeling an onion. We hit the first layer. And we knocked that M'er F'er out of the M'er F'ing park. From this point forward they will KNOW community feedback is more than just the rhetorical whining of someone who doesn't know their rotations getting beasted on a scoreboard.
    • Like x 15
  13. Rokyn Dedicated Player



    When you test stuff with that method it doesn't give you the complete picture per se. It's like doing a report on some project but rather than break it down individually with data and large sample sizes....a person would instead do very surface level testing with many variables to account for and if someone wants to repeat your test then they will have trouble reproducing a similar test in order to give further evidence or disprove it. That's what scientific reports are based around is if it's reproducible/testable through experiments, number of variables used, sample sizes, and etc. The easier and simple you make the tests and stats, the better it is to be reproduced.

    We don't want to force the devs to really adapt to our level when they know much more mathematically wise about the game than the majority of players do. It's actually quite refreshing they are giving us a method to test by since it will produce results that can be reproduced by them which is what we want so they can see why x or z needs to be brought up in damage or so or whatever. And if there is something that needs to be pointed across to the devs in terms of for example such as x power does less here and there or the concept of how unfair 12.5 meters can be to most powers in game.....than these tests will probably be what they observe and recognize.

    Right now we were given a method to use and if this method means we can get our points across to the developers that they can actually consider and reproduce them, then the better it is for us. Work hand in hand with the developers, don't try to belittle their work. You can disagree and backup your opinion which is perfectly fine but you must have the incentive to try to see their point of view first and then come up with a testable hypothesis after data collecting.

    Also, the tests you are advocating about that include every possible way of damage are actually just a higher scalar value than the Tunso tests as proven by Sore. There's a pattern there and so it's not really about how to get the highest parse or maximize your numbers but to really understand what's going on in each individual broken-down 10 seconds. The more samples the better and the 10 seconds does this. The less variables the better for reproducing and having a similar stat line allows for fair testing among all power sets.
    • Like x 3
  14. CrappyHeals Devoted Player

    I know its not about the highest parse I just personally dont like the 10s cause of the highs and the lows. You base stuff mostly on averages but its nice to know the peak and the bottom and with the 10s you get false peaks and its more random. I know i know you can do some math and get the average but why just not do 30s and get consistent data. I know perfectly well how to test things and how things work i don't need spread sheets and graphs this is not rocket science, my lord people lol
  15. Crimson Mayhem Loyal Player

    Great speech man. I kept thinking it would end on something like
    Sore4President
    • Like x 2
  16. Fies Committed Player

    I would prefer to go like this:

    a) Try hard to execute the "perfect rotation"->capture input->script->run script 30min+ (human failure and randomness excluded)
    Compare. -> adjust
    Important: Every Powerset needs atleast 1 viable range ( >/=12.5) rotation.

    b) Find "factors" for at least the following: 1) build-up time (burst) 2) interuptibility (combo-powers/"AM-window") 3) ease of use 4) range [5) mobility] These can be decided according to experience and modified later on. Factors represent "ingame-conditions".

    ("bad") example with made up numbers/factors:
    ---------------------------"burst"---"interupt."----"ease"--"range"--"mobility"
    ICE 1kDmg/sec-----x 1---------x 1-------------x 1------x 1,5--------x 1------ = 1500 Dmg/sec
    ELE 1kDmg/sec----x 1,3------x 1------------x 1,2-----x 1,0-------x 1,1---- = 1716 Dmg/sec
    => 1500 Dmg/sec "Ice" equals 1716 Dmg/sec "Elec" under ingame conditions=>"balance"
    Again: bad example/numbers!

    c) Adjust Dmg accordingly -> Test under ingame conditions (raids/SM..)
  17. Pults Loyal Player

    I did rage testing using a macro. Although it was done prior to Tunso's announced testing standards. It completely ignored any human made errors and it gave somewhat even results in all outcomes.
    • Like x 1
  18. Radium Devoted Player

    Not even gunna lie, I had to take a half hour break between testing Rage Melee because my hands were that Sore. Playing Ice AM has made me soft and weak when it comes to clipping.
    • Like x 4
  19. Pults Loyal Player

    Worst part is one must do insane clipping to even compete with melee whilst other powers simply have to be within their range and slowly click one button at a time. Fire and ice can do it with just one button.
    Rage without clipping should be in the 12-14k melee range.
  20. Immortal Kyrro Loyal Player

    I get the sanme thing with sorcery but I can't compete just try not to get beat by the crap dps.
Thread Status:
Not open for further replies.