+ Reply to Thread
Results 1 to 1 of 1

Thread: What are missing from dummy parsing?

  1. #1
    Soulwalker
    Join Date
    Jul 2013
    Posts
    11

    Default What are missing from dummy parsing?

    Maybe a list of factors that are not taken into account during dummy parsing would be a help to the developers? I am sure they already have such a list, but if we put one together they have something to double check against?

    We might even come up with some values for the differences that could be used for comparison between souls.

    I don't feel qualified to put such a list together and even less so about the actual values, but I can give you a few examples that has been brought up again and agian in different threads:

    - Equipment scaling with BiS vs. PTS gear. Should be different between all souls but mainly stay the same regardless of rotations and minor details.
    One example could be that if Shamans BiS itemization gear on a normal server is 10% better than a PTS equivalent itemization (with and without crit power) then Shamans get a coefficient of 1.10 for this. While another soul might have an optimal itemization on the PTS gear and get a 1.00 for their koefficient.

    - Difficulty level of rotation. How much deviation do a good player have from the optimal rotation during actual play? Some classes are more forgiving than others which means that the harder classes should have a little premium to make up for that.
    Here we will have to pretend and compare optimal rotation and then throw in a few mistakes and se how that compares to another soul's optimal and suboptimal parses. Then add a little guesswork of how often the mistakes will be made. I would guess this can only be guestimated by people who play more than one class at a raid level. Or at least raid regularily with more than one soul.

    - General survivability. Since a dead dps doesn't really do much dps when in a vertical postion.
    Here one could set up a general squishyness index in a raid. This means we should check logs and see what souls tend to die most often? If we start to compare actual abilities we will have a huge amount of data. But saying that Harbingers spend on average 2% of the raid time on their back would yield a coefficient of 0.98. Here we need a lot of comparisons from different guilds or official servers stats. Only successfull boss fights should be included of course.

    - Abilites not used in dummy parsing like Rage Blight and Glacial Strike that only comes in during parts of a fight or not at all in a dummy parsing.
    So if Rage Blight has an uptime of 80% and increase damage by 5% then that ability in itself gives a koefficient of 1.04. Here we should be able to produce the most complete list of coefficients and probably the most accurate.

    - Target switching and disconnect factors.
    Mostly a negative one for most classes I would guess. Here we need two values one that is shared by everyone which is how often there are switching and disconnects on average and then one for each soul that shows how much damage goes down when it kicks in. If we guestimate the time factor to be 20% and that the soul loses 20% damage during a disconnect or target switch the coefficient would be 0.96. It might also be useful to split switching and disconnects up into two diifferent categories.

    - Mobility
    How much time do we generally have to dps on the move and how much do each soul's dps drop when they have to dps on the move? Important that we do not mix this with disconnects in general. This is more a measure of channeling and long cast time problems.


    - AOE potential and maybe 2 or 3 target potential. One could probably calculate how much time is spent in such situations and just add X% of the AOE damage increase to the total.This one is a little tricky and it might be good to split it up in 3 different categories. The important thing is to multiply the time with the dps increase. If you spend 5% of the time doing 100k dps instead of the normal 20k your coefficient would be 1.2.

    My guess is that most of the above coefficients would be in the order of 1.05 and multiplied together they would be fairly equal for most classes with some better at one thing and worse in other areas. Melee should score less and therefor deserve the current dps premium they enjoy. But one could also find out that certain classes needs to perform significantly higher or lower on dummies in order to be balanced in a real PvE environment.

    Lets put a serious list together so the developers can double check their own list (if they actually have such a list explicitly written out, they might have relied mostly on parses themselves so far and a really good knowledge of the game.) And if we can produce some guestimate values for the different souls as well it would be even better.

    One good thing is that once we have this list we could save alot of discussions and disagreements about "other" factors. And the list would most likely stay pretty much the same for most souls for some time. It takes alot of tweaks to change itemization coefficients or general survivability factors. And if we are lucky the developers might share their list, thoughts and coefficients with us.
    Last edited by Quirky1; 07-28-2013 at 04:23 PM.

+ Reply to Thread

Bookmarks

Posting Permissions

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