Results -9 to 0 of 521

Threaded View

  1. #10
    Player
    Ryel's Avatar
    Join Date
    Sep 2011
    Posts
    610
    Character
    Ryel Altaria
    World
    Excalibur
    Main Class
    Samurai Lv 80
    Quote Originally Posted by UBERHAXED View Post
    Why do you need to know (example here) that you are performing below the average dragoon by 300 dps before (completely arbitrary value) you decide you need to play correctly. If you are playing at 100%, not missing any GCD or positionals and performing the standard burst at the dps check, then there is NOTHING you can do to improve your DPS, other than rely on random variables. If you know you are not playing at 100% (that is, you are missing some GCDs, your cooldowns were not aligned for the dps check, you missed positionals) then you already know, without seeing any value that you are performing subpar and you can improve.
    (Example incoming)

    It's like saying that if you miss bunch of positionals (due to tack positioning) in equal concentration it's okay if you are only "30 DPS" below the average. But what if the average Dragoon also doesn't clear the fight, due to them not being good players (not exactly a stretch here). All of the slower groups who don't form their strategy uses the same meta strategy and positions the boss in the same terrible way because they saw some guide video of doing it that way.
    I think this is where you're misunderstanding.

    In this example it's not a matter of the player making a conscious choice to play below the average Dragoon by 300 DPS it's the fact that they may not be fully aware they are doing so and the comparison by parser begins to illuminate any issues with their play. The issue is what the player believes to be their 100% might turn out isn't the case, that they weren't aware they were having issues or making mistakes.

    Stop me if you've heard this one before: "Oh man, I thought my DPS should have been higher than that, I know I can do better than the last run, let's run it again!"

    Seeing tangible feedback allows a player to evaluate themselves in an effort to improve, nobody is saying the parser is going to do the work for you but it will show you the results of the current attempt. If it was the case that every single player had full awareness of their mechanics executions and rotations on equal grounds and equal fields there would be very few poor players across the board, however this has proven to be not the case. This concept isn't even exclusive to the game it applies to everything.

    in fact we have multiple sayings for it:

    "Practice makes perfect"
    "If at first you don't succeed, try again"

    Getting feedback on how your rotation actually handles in live content versus the vacuum of a dummy parse is the only data that actually matters, because it's when it counts. You must have seen various comments in the DPS forums / reddit / anywhere of players asking others how they've gotten certain results or seen videos of people showing off opening rotations, speed kills etc because actual player skill is a factor not just the rotation on it's own.

    Parse results are key feedback when it comes to refining your skill.

    Quote Originally Posted by UBERHAXED View Post
    But instead of actually solving the problem (telling your tank that the boss positioning needs to change, regardless of how they saw the tank do it on the video so you can do your positionals) you are content because you are only 2% behind the average. you knew for a fact that you can easily improve your dps by doing something so fundamental, but were willing to accept being subpar is okay as long as everyone is also subpar. This reasoning is why there is a large gap between the average player, and player who have cleared the raid tier. And guess what? If you want a parser, it's probably because you intend on clearing the raid tier.
    Nobody said this wasn't part of it, it's not like people are saying you'll look at parse results and instantly become a top tier player and never have to communicate with your party.

    A lot of this conversation actually takes place when the party looks at the parser after a failed DPS check, sees they might need to squeeze in some more DPS in various areas and says:

    "Okay guys what can we do here?"

    This is usually followed by an analysis of what can be changed on everyone's end, things like: Where can a healer get in some more damage? What can the tanks do to help the DPS? etc etc

    Party assessment has to start somewhere, having tangible results you can point to and say "we did this well versus this well" is a good starting point for improvement.

    Quote Originally Posted by UBERHAXED View Post
    The fact that you were doing 300 behind average even if your rotation was optimal (as in, your damage cannot improve) might stem from the fact that you were using a 170 weapon. The parser will say the same thing, but you will now ask around what other dragoons do instead of seeing, "Obviously, this is already striking dummy burst rotation. It can't be improved mathematically, it must be the 10 points of weapon damage I am missing". If you're not doing perfect striking dummy burst rotation, then you already know (without a parser) that you can improve your DPS and exactly how to improve your DPS.
    (Example based on example incoming)
    It's like saying you are taking an engineering class of 300 people.
    ->
    The average grade is a C and your grade is a C. You then come to the conclusion that your are doing it right.
    ->
    The average grade is a D and your grade is a C. You then come to the conclusion that you are doing it right and also don't need to work any harder because you are already better than average.
    ->The average is a B and your grade is a C. You then come to the conclusion that you should should spend another hour a week to review your notes before exams because you are slightly below average.

    The issue is that none of these scenarios are correct. Knowledge of what others are doing should not have influenced your decisions to do better or not do better. If your grade was a C you know from the start that you need to improve because you haven't hit the maximum (A+). Similarly, Even if you didn't have knowledge of the numerical value of your grade, you know that you missed some test questions and you can always improve until thew point that you have perfect accuracy on every topic.
    These are all things you evaluate and assess based on feedback.

    No offense but you make comments that lead me to wonder if you've ever interacted with another person and i don't mean that as an insult.

    Are you saying you've never come across a situation where one or a number of people were inspired or encouraged to perform better based on the results of their peers? You've never seen someone look at the class average of test results and think "man next time i want to get an A instead of the average of a C" while some are perfectly happy thinking "well everyone else got a C so i guess I'm not doing all that bad, so why bother trying any harder?"

    In the case of a lack of a parser it's like taking a test with 300 people and instead of getting a grade back at all the professor just stands up and says "You all passed" or "Most of you passed, some of you failed, but I'm passing the entire class anyway because the majority of you did well enough that i don't care about the rest" it's like "Okay, well that's great and all but, I'd really like to know how i did on the test"

    Of course this whole testing analogy is terrible to begin with because tests normally aren't dynamic and don't change questions during the exam nor are they effected by the actions or lack of actions of other people taking the test at the same time, that and when it comes to mathematics there usually isn't a physical component involved in taking a test (rotation, practice, etc).

    However it's like you're trying to remove factors like motivation, practice, and a desire for improvement from the conversation as if they aren't very real reasons as to why some players get better and others don't. Regardless of your personal feelings on the matter, the evidence (see: human history) speaks otherwise.

    There are plenty of people who should be better at things that they do but they aren't, there are other people who get better at things when they have a form of measurement placed in front of them showing how well they should be doing and use it as a basis for improvement.

    The argument isn't about improving the mathematics it's about improving the execution, which in turn is rewarded with better results.

    Quote Originally Posted by UBERHAXED View Post
    The amount of misinformation is disturbing. The challenger had a defect that was known to occur ahead of time due to weather conditions (physics at play here). The shuttle was launched due to scheduling constraints, leading to the situation that it had to be launched that day, or some time in the following year. This was a classic example of management ignoring engineering information that was known as result of equations beforehand before the launch. If nothing else, this further augments the point. "Well, the math could be wrong and the seal on the rings don't break" was the "flawed" argument here. It would be in your best interest not to discuss what you have not clue about.
    Don't want to delve too deep into the challenger thing but the point wasn't to disprove the mathematics, but to illustrate there are factors beyond them.

    When i say nobody saw it (a malfunction) coming i mean nobody saw the scope of the disaster coming. "This launch might backfire" and "Everyone on board is going to die" are two very different things.

    Were there reports since the 70's of the O-ring flaws and the lack of data for various temperature ratings as well as the fact that this known issue hadn't been resolved? Yep.

    Were there multiple warnings from the engineers that something could possibly go wrong and they weren't 100% on whether the launch would have gone as expected, with in fact data saying it might not at all? Yep.

    Did the higher-ups ignore or not receive this feedback at all this based on a combination of launch scheduling and others feeling they were safe enough to do so anyway? Yep.

    And did it quite literally explode in their faces? Oh yeah.

    Had they known for 100% what the results would have been would they have done it anyway? /shrug who knows?

    My point primarily is that while a rotation says you should be performing in one way (with a degree of variance) a player's skill and practice comes in to play when seeing if these results line up.

    I'm also not entirely sure why you seem so hostile over this either.
    (12)
    Last edited by Ryel; 12-05-2015 at 08:18 PM.