1. #1

    Pristine Protoscale Girdle in 7.1: Talent choices?

    So I just picked up pristine proto-scale girdle last night to go with my ring of the twisting nether. Since we are still in 7.1, I thought I would sim taking elemental fusion over primal elementalist. All the conventional wisdom I've seen indicated that it wasn't worth changing talents.

    However, using the Mr. Robot sim tool, I'm finding that I would get an extra 5k dps out of elemental fusion over primal elementalist. Granted that ends up being a 1% difference, but it's not a complicated talent to use and 1% is better than 0%.

    Did I make a mistake? Or is Mr. Robot messed up? Anyone else try this out?

  2. #2
    A. Nobody missed something this huge after 4 months of the expac being out.
    B. Stop using AMR, it's the poor man's Simcraft.

  3. #3
    Hey Rokkshark, I'm from Ask Mr. Robot and can fill you in on a few things.

    1. We test our sim results against real logs, using the Ursoc script. So far it is spot on, including elemental. It's a nice way for us to make sure that the sims are being realistic, and they are definitely for DPS (easier to measure than tanking/healing).

    2. Every character performs differently, based on the gear they have. So anyone telling you one thing is always better than another without taking a look at your gear is giving general advice. It might end up right, and is probably right most of the time, but not every time.

    @Drarakaun - The AMR simulator is built to be really advanced with more powerful features. If you haven't played with it much, here's a few things that we have available for people to use:
    - Batch sims: f you want to run a batch sim, we make all of the combinations for you (often hundreds or thousands), so you don't have to do it by hand like in SimC.
    - Stat weights: We look at 1000s of data points (instead of just 2 like simC) and include stat interactions (unlike SimC), which gives you more statistically sound stat weights. - - Sim reports: We generate full searchable logs so you can figure out what is happening (and find bugs too). We also keep a full history of all of your sims so you can refer to them later.
    - Sim models: We have a more advanced timing model that you can configure super easily.

    There's more, but that's a start. If you can't find a feature you're looking for, let me know. I will either show you how to use it, or if we don't have it, add it to our list
    Ask Mr. Robot Human Minion

  4. #4
    Sure enough my results as follows:

    Ask Mr. Robot:
    Primal Elementalist: 420k
    Elemental Fusion: 425k

    SimulationCraft:
    Primal Elementalist: 430k
    Elemental Fusion: 417k


    So why the disparity? @Drarakaun what makes simcraft inherently better than mr. robot?

  5. #5
    Deleted
    It's pretty simple, if AMR sims EF higher than PE it means that AMR has no idea what it's doing and should be disregarded as far as your character is concerned.

    PE is SIGNIFICANTLY ahead of EF even with Belt.

  6. #6
    I always laugh at people that simcraft, can't you all just play the game and learn how to play.

    You can sim all day I'll still fucking destroy you in performance because I know how to read my skills and figure out the correct combination of when to hit spells because I have a brain.

    Simming is unrealistic and shit.

    In laymans terms: GET GUD.

  7. #7
    Quote Originally Posted by Funky View Post
    I always laugh at people that simcraft, can't you all just play the game and learn how to play.

    You can sim all day I'll still fucking destroy you in performance because I know how to read my skills and figure out the correct combination of when to hit spells because I have a brain.

    Simming is unrealistic and shit.

    In laymans terms: GET GUD.
    Wanna post some logs? Or nah just git gud.

  8. #8
    Bloodsail Admiral kosajk's Avatar
    15+ Year Old Account
    Join Date
    Jul 2008
    Location
    PL
    Posts
    1,145
    Quote Originally Posted by Funky View Post
    I always laugh at people that simcraft, can't you all just play the game and learn how to play.

    You can sim all day I'll still fucking destroy you in performance because I know how to read my skills and figure out the correct combination of when to hit spells because I have a brain.

    Simming is unrealistic and shit.

    In laymans terms: GET GUD.
    ok i must admit i don't get you, simulationcraft and amr are just a tools you can use them or not and play good or bad with or without them, so if you get your information from forums or wowhead for instance you are as bad as anyone using simcraft cause its just another tool to get your info, based on your comment only proper way to be a good player is testing every single thing by yourself for hours, gg with it

    tbh i don't find anything bad in wanting to know if one gear/talent combination is better than another especially with current trinkets or gear with special bonuses, where sometimes they can make a big differences and really some in game tool-tips are not reflecting the super potential of given item for one class/spec while at the same time being medicore or just bad for another, not mentioning titan forged items

    ps. siming thing takes 1 relog (~30 sec) then importing your char (like 3 clicks ~15 sec) and then after 2-5 mins comparing results (~10 sec) so ~1 min of your time to be 99,90% sure that item just drooped/talent choice is upgrade or downgrade in your case, seams like really stupid thing to do idd Oo

    "Hope for the best and prepare for the worst"

  9. #9
    @rokkshark - if you have links to the reports, that will help me triple check things and do some digging. I'll hit up our theorycrafter too, have it take a look and compare it to logs as well.
    Ask Mr. Robot Human Minion

  10. #10
    Quote Originally Posted by Zoopercat View Post
    @rokkshark - if you have links to the reports, that will help me triple check things and do some digging. I'll hit up our theorycrafter too, have it take a look and compare it to logs as well.
    @Zoopercat
    Using my character: Rokksteady @KILRogg-US

    With Primal Elementalist:
    http://www.askmrrobot.com/wow/simula...1cd7bda4cd376d

    With Elemental Fusion:
    http://www.askmrrobot.com/wow/simula...b9dfac2e52257c

    If you can figure out what is going on that would be great. I vastly prefer the interface and extra options in the Mr. robot sim, but all the theorycrafters seem to be in agreement that even with the belt Elemental Fusion is far beneath Primal Elementalist.

  11. #11
    The belt was nerfed on ptr to be the same as it is on live, it doesn't change the talent build anything

    No matter what legends u have
    On live only 1 build is viable and that is with PE
    On ptr only 1 build is viable and that is with icefury and ele blast

    Don't use AMR to sim trinkets or talents use sim craft for that
    AMR is def ok for gear sims tho, I've never had any dps problems going with their gear stats value

  12. #12
    @rokkshark I did a bunch of tests, and it looks like yes, Elemental Fusion might be just as good, or better than Primal Elementalist because of your legendary. That legendary increases the dmg of lava burst, so that helps close the gap on the EF talent, when PE is normally better.
    Last edited by Zoopercat; 2016-12-20 at 06:46 PM.
    Ask Mr. Robot Human Minion

  13. #13
    Quote Originally Posted by Spoonman View Post
    On ptr only 1 build is viable and that is with icefury and ele blast
    This is not true. Ascendance is perfectly viable and competitive in real raiding environments where playing IF/EB to perfection is impossible, not to mention in a council situation where asc pretty much blows IF away when you take EF.

    Source: I'm Claptrap, I did the theorycrafting with slander and bloodmallet.

  14. #14
    @Zoopercat I looked through the Mr. Robot sim and it looks like the sim applies the DoT on lavaburst AND lavaburst overloads. This is incorrect. The DoT only works on the actual spell, not overloads.

    I see under the Pristine Proto-scale girdle in the item subsection of the Live Version:
    Triggers: Type: Filter: Whitelist: LavaBurst, LavaBurstOverload

    This only accounts for 1-2k dps though so still doesn't completely resolve the discrepancy between Mr. Robot and other sims.

    I found that the source of the rest of the damage is from your lava burst calculation. Mr. Robot lavabursts with the the same rotation, stats and other variables does about 10% more damage with each cast of lavaburst than SimC. I've looked at the spell information for both Ask Mr. Robot and SimC and I can't find the source for this discrepancy. But with all the same variables, your lava burst always hits harder.

    - - - Updated - - -
    @Zoopercat I think I found it!

    It looks like the Molten Blast artifact trait calculation in Ask Mr. Robot is wrong. The trait adds 2% critical strike damage per rank on live. Your sim adds 4% critical strike damage per rank.

    I have a relic for that trait, meaning Ask Mr. Robot is giving me 8% more lava burst critical strike damage than it should.

    See here:
    http://www.askmrrobot.com/wow/theory...l&version=live
    @Zoopercat if you guys could fix those two bugs I think your sim will be back on track.

    In the mean time I can make a custom setup for that spell.

  15. #15
    @rokkshark - you are my new favorite person for the day! Thanks for doing a thorough comparison.

    Edit: here are some updates after talking to our TCer and dev. The crit multiplier is actually correct. I know it seems like it isn't, but here is some documentation to point you to: http://www.askmrrobot.com/wow/simulator/docs/critdamage

    TLDR: there are 2 kinds of crit multipliers. If you look in the spell data for that trait it is 4%, but the tooltip shows 2%. I know, confusing.

    As for the item - checking into that too. I think you're right on this one and there isn't anything special going on. Do you have any logs or anything I can review just to be ultra-super-duper sure?

    <3
    Last edited by Zoopercat; 2016-12-21 at 03:32 AM.
    Ask Mr. Robot Human Minion

  16. #16
    @Zoopercat

    Here is a report using the live version, a simple rotation, and no flasks/pots/food:
    http://www.askmrrobot.com/wow/simula...6a31d784fc0c0f

    And here is the same test using the custom version I have with just the Belt proc fix for LvB overload:
    http://www.askmrrobot.com/wow/simula...64f2dacecfd5bf

    It's a small difference but it's there. The live version has the proc doing .5% more damage than with the fix. I just confirmed in game that when overloading you only get one stack of the DoT.


    I still haven't figured out the rest of the discrepancy with the lavaburst damage. I'll have to dive a bit deeper.
    Last edited by rokkshark; 2016-12-21 at 05:33 AM.

  17. #17
    @rokkshark - thanks, we updated the girdle on our dev server. It will be live with our next update.
    Ask Mr. Robot Human Minion

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •