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. Black Jaq Devoted Player

    the inherent fallacy behind the diagrams that Sore did is that for Ice and Fire who use the short range attacks, they gain no damage past the maximum range of the ability, unlike Quantum, for example, which does have the potential of hitting targets behind the 12.5m intended target. And truly, if you are fighting a single target boss, there is no damage to consider past the intended target.

    I plan on also testing Electricity and Earth if I have time this weekend.
    • Like x 2
  2. Remander Steadfast Player

    Agree, shiny. The agnostic testing assumes that the powersets will scale equally with modding, SP allocation, and increasing crit chance/mag. I doubt that's accurate, and it likely explains the disconnect between the players and Devs, when it comes to balance.
    • Like x 7
  3. TheDark Devoted Player

    I also use 1 minute parsers but with 10 to 30 samples. 10 second parsers are just shorter and leave room to throw out parts of the rotation you mess up. In a 1 minute parser, if you make a mistake, you waste time and have to throw away that sample. Some can do rotations perfect and some just run into bloopers occasionally.

    I will continue to use 1 minute parsers to compare loadouts within my powersets but I will also use these "Tunso Tests" To compare with other powersets. Thing is a majority of players don't know how to test. Popular YouTube or twitch players especially. Everyone throws in all these variables and they never inform viewers or readers of every single detail they had when they tested like stats, HT mods, skill point amount etc. "Tunso Tests" is just a template and provide a guide for everyone to follow that HE will acknowledge. The devs have acknowledged other ways of testing too but under the circumstances that everything used was included in the description of the rotation.

    Bugs will still be able to get found by few. Players have to test powers individually which is currently not done. Sometimes, players are oblivious to bugs anyway. You just reduce variables by excluding HT.

    Testing the same way is ideal under certain circumstances. Why compare tests to someone who used range powers with HT to a person who used melee without HT and modded for crit bonuses? The tests are more controlled this way. I don't agree 100% with a few of the things mentioned in Tunso's post. ultimately we have to play by his rules to send messages. This is something he should have mentioned at the very beginning of the AM release imo.
    • Like x 2
  4. Sage-Rapha Steadfast Player

    My sub expired on pc, however, if you guys can test this:
    WM Sorc vs Sorc AM with a Gadgets troll using Battle Drone. (That's just for a record of potential simulating a piece of content.)
    If not, just continue despawns
    ten second interval same spec.

    And yes it may appear that way in the Tunso spec, as I get 20k using despawns live. Leaving pet out, 17k. However in a raid setting, it's not good without the battle drone or despawning. And that should not be the case. I may be doing something wrong, and If so, I own up to it.
    But WM can outdo that and that's the trouble. There's no way of beating WM without the pets seeming overpowered.
    I would love for offering give a 40% boost to the two powers used after it in the summoning and destiny trees.
    copy and paste for Earth on Fortify Golem, Geokinesis and Seismic trees.
    Now of course they gotta use weapons.. but anyway.
  5. Xibo Loyal Player

    Exactly, but how we could solve this? Light Claws, for example, ONLY can be execute in 7m how you say. My suggestion would be become melee the range attacks (25m) for all powers and eliminate the "risk/reward" option. So you can become balanced all the powers attacks doing melee damage or range damage regardless of the distance.

    The devs can establish a pattern damage for all the AM's and apply him for all the powers regardless of distance or ability.
  6. Sore Steadfast Player

    It seems some of my Nature tests scale around x1.4 to live. Ice scales close to x1.5 to live. I think one reasonable explanation is the pet-factor. In a Tunso Test, Swarm does CR-based damage and won't scale any further on live. That reduces how much to scale a Tunso Test to live.

    I know the thinned out tests creates these kinds of variations. The hope is that the difference is only an acceptable variation or something we'll be able to tease out when we give feedback in the future (like x1.4 vs. x1.5). Obviously there was a need for a standard testing approach. We saw some gross variation in what players were doing in our threads. This approach was to reduce the barrier to entry for testing. Quicker to setup toons.
  7. CrappyHeals Devoted Player

    I just think they need to set a more realistic testing standard lets say all the crits in the trees, the 2 dps innates, correct modding, max damage,core strength and enough skill points to fill out 2 weapon mastery trees and the dps innates at the bottom incase you have to test something with a WM set up.

    Not modding gear leaves out some crits and that can be a big variable on some power sets not to mention other things like buff% and other things.

    Did you ever get done testing with with modded gear?
  8. Radium Devoted Player

    Thats where the 1.4 vs 1.5 is happening. I scale better compared to Sores Nature and we have it nailed down to Swarm being the reason. Its possible theres something else going on so I'm going to check Fire and Mental to see if the damage scales the same for them compared to Ice.
    • Like x 1
  9. Remander Steadfast Player

    Good point about pets. Perhaps, pets should scale with stats, rather than CR.
    • Like x 4
  10. Sore Steadfast Player

    I've been given the okay to share that these are the targets we are working with in a Tunso Test.

    With the setup Tunso describes, our targets (10 second average):
    For Range: 14120
    For Melee: 20644

    Currently they are okay with a +/- 5% variation. That said, our efforts in making Tunso Test to Live comparisons is whether we should tweak those targets for specific powersets. The answer is clearly yes looking at Ice vs. Nature because of the CR-based pet. I want to do a few more Nature tests to sharpen the pencil on that scaling. But more so, we just need more test results from other powersets now.
    • Like x 3
  11. Sore Steadfast Player

    As for scaling, I'm also planning on seeing if I can predict scaling within a reasonable tolerance for non-pet classes just based on the proportions might and precision increase between a Tunso Test and Live. I know that doesn't cover other factors (core strength, affinities, etc) but maybe we can get a model that's not 100% mathematically matching systems but get a model that's statistically predictive.
  12. CrappyHeals Devoted Player

    I just feel like this is a lot of extra work for nothing and hats off to you guys doing the leg work but if the devs just changed their methods we'd be fine. Thats just my opinion though.
    • Like x 1
  13. Sage-Rapha Steadfast Player

    This right here..

    Though Im curious to see what Radium is getting with Ice
  14. Radium Devoted Player

    20k average base, 31k average with everything
    • Like x 3
  15. Radium Devoted Player

    Part of this is finding flaws, put its also discovering the disconnect between what the devs see and what we see. At least now we can physically show them "What you see and what we see are not the same."
    • Like x 5
  16. Sage-Rapha Steadfast Player

    Wow. No Kidding lol. Ice is fast as hell
    That's kicking Earth and Sorcery's ***.
  17. Radium Devoted Player

    Like the Nature comparison, I suspect when Earth and Sorcery get tested they'll show much closer to what everyone is at using Tunsos method. But because the pets account for far more damage, you'll see a larger disparity on Live tests compared to Nature.
  18. Sage-Rapha Steadfast Player

    Unfortunately that's the case from what I see.
    Im not going to post my numbers yet but I will soon. :)
    It's simply due to Fury doing like 2/3 of my damage.
  19. Sore Steadfast Player

    It might be more complex that just the pet scaling. It might get more to what Shiny was hinting at with crit scaling too.

    Nature Melee with Swarm: Scaled 1.34-1.41
    Nature Melee with Voracious Plants: Scaled 1.21-1.27 (it was also a very weak rotation)
    Wolf/Dog Melee: Scaled: 1.47-1.56
    Human Mid-Range with Swarm/Carnage: Scaled 1.30-1.39
    Human->Wolf Mid-Range: Scaled 1.40-1.43

    I know others have really been harping on this point and I'm coming around to it. I'm really starting to think it's too erratic to spec the toon like this. I'm mostly worried about not having Core Strength, Escalating Might/Relentless Precision, Exobyte Mods, and Affinities. I'm thinking League Proficiencies and Generator Mods might not be as important. But it is good that we're taking the time to prove with results. I'd still like to see what people see with other powersets because I don't want to draw conclusions from just one.
    • Like x 4
  20. Electrizzaro Loyal Player

    Do you have any vids of how you are doing the tests? Reading how they are being done is one thing but seeing it is another. It would also give me a better idea of how everything is fitting together.
Thread Status:
Not open for further replies.