Page 7 of 7 FirstFirst ... 5 6 7
Results 61 to 70 of 70
  1. #61
    Player
    Rongway's Avatar
    Join Date
    Aug 2013
    Posts
    4,147
    Character
    Cyrillo Rongway
    World
    Hyperion
    Main Class
    Black Mage Lv 100
    Quote Originally Posted by VictorSpoils View Post
    • BLM has long casts, random procs, the worst survivability, the worst weaving, the worst mobility, etc. I still don't like this job.
    It's fine if you dislike it, and you don't even need a reason for doing so. But at high levels BLM has reliably forced procs and at least 30% full mobility from instant casts and also access to short cooldown teleports. This makes them more on-demand mobile than RDM, whose long duration mobility is only useful if the movement allows 5s in melee range and requires a ramp-up of 8 dualcasts (40s). BLM also has a lossless shield, which makes them more survivable than RDM in high level practical applications.
    (5)
    Last edited by Rongway; 12-05-2022 at 09:36 PM.
    Error 3102 Club, Order of the 52nd Hour

  2. #62
    Player
    Valence's Avatar
    Join Date
    Oct 2018
    Posts
    3,731
    Character
    Sunie Dakwhil
    World
    Twintania
    Main Class
    Machinist Lv 100
    People conveniently seem to always forget about Enchanted Reprise. But oh -gasp- it's suboptimal! Well yeah, essentially like most mobility tools, and this one isn't that shabby. You also have a lot of suboptimal uses for BLM mobility tools, but people don't want to acknowledge those (triple cast and nukes out of buff windows among others).
    (1)

  3. #63
    Player
    Azuri's Avatar
    Join Date
    Dec 2021
    Posts
    769
    Character
    Azuri Aeru
    World
    Phoenix
    Main Class
    Red Mage Lv 90
    Quote Originally Posted by Valence View Post
    People conveniently seem to always forget about Enchanted Reprise. But oh -gasp- it's suboptimal! Well yeah, essentially like most mobility tools, and this one isn't that shabby. You also have a lot of suboptimal uses for BLM mobility tools, but people don't want to acknowledge those (triple cast and nukes out of buff windows among others).
    BLM has Scathe. It's about on par with Enchanted Reprise in usefulness.
    (4)

  4. #64
    Player Mithron's Avatar
    Join Date
    Dec 2016
    Location
    Ul'dah
    Posts
    699
    Character
    Mithron Scarlet
    World
    Gilgamesh
    Main Class
    Paladin Lv 100
    Quote Originally Posted by Valence View Post
    People conveniently seem to always forget about Enchanted Reprise. But oh -gasp- it's suboptimal! Well yeah, essentially like most mobility tools, and this one isn't that shabby. You also have a lot of suboptimal uses for BLM mobility tools, but people don't want to acknowledge those (triple cast and nukes out of buff windows among others).
    It's unusable at 5 Mana cost.
    (3)

  5. #65
    Player
    Shiroe's Avatar
    Join Date
    Apr 2014
    Posts
    866
    Character
    Ohlala Chica
    World
    Zodiark
    Main Class
    Alchemist Lv 100
    Quote Originally Posted by Tsumdere View Post
    It's not objective and you can't glean much information in the way of difficulty from variance. There are other factors that play into why a job has a high/low variance.

    Variance can be altered by number of parses (DNC has triple the parses of MCH), influence of crits on the job, flat out mathematical damage potential capping the ceiling, job responsibility (did you know RDM has a higher variance than all melee but DRG, sometimes the second highest variance behind BLM depending on the fight), and the fight itself.

    -

    If we are to take damage variance as the actual difficulty of classes (based on savage), it would be:

    BLM > BRD > DRG > RDM > DNC > RPR > MNK > NIN > SMN > SAM > MCH

    And for fun so you can see the affects of how it can change drastically per encounter:

    DSR: DRG > BLM > MNK > SMN > SAM > DNC > RPR > BRD > MCH > RDM > NIN
    Criterion: MCH > SMN > SAM > DRG > BLM > RPR > NIN > DNC > MNK > RDM > BRD
    TEA: BLM > RDM > NIN > DRG > SAM > DNC > MCH > MNK > SMN > RPR > BRD

    It's kind of all over the place with only two consistent factors (DRG/BLM in top 4). If we were to take the data at face value - SMN is harder than SAM, DNC is harder than MNK, NIN is a melee for brainlets, and DRG shitstomps over the other melees in difficulty.
    ^ this
    (thank you for the post)
    ....

    Smn is braindead on lower levels, and even lv 90 dungeons.. , but on other content
    (1)

  6. #66
    Player
    GutsyBoozetank's Avatar
    Join Date
    Mar 2015
    Posts
    24
    Character
    Mio Mina
    World
    Balmung
    Main Class
    Alchemist Lv 71
    I love that the class I've mained since ARR is considered one of the most difficult classes. Since playing it for so long my opinion is reversed and I feel like every job but BLM is the most difficult lol
    (1)

  7. #67
    Player
    Renathras's Avatar
    Join Date
    Dec 2014
    Posts
    2,747
    Character
    Ren Thras
    World
    Famfrit
    Main Class
    White Mage Lv 100
    I think the most objective measure would be to use the forbidden sight and look at how wide the bar is from lowest 1%er to highest 99%er. The bar denotes the spread of damage output, meaning "what the least skilled output vs what the most skilled output". A wide bar means that skill/experience makes a huge difference, which heavily implies that the Job is more difficult to play and master. A narrow bar heavily implies that the Job is easier to master. While one can argue this isn't a complete picture - no metric WILL be, mind you - it's probably one of the best ones we have. It doesn't take into account "other things" (like Healers outputting healing, or how easy a Job is to first pick up and "do passably with", and so on), but it's still a good way of measuring within roles and across the game. We can show this by testing a few examples.

    DISCLAIMER: The following are for instructional purposes. I generally hate this site and don't run the program myself (and if you look me up, you'd see I don't often run hard content to bother anyway), but just to give you an idea.



    WAR has the narrowest band of the Tanks, SGE of the Healers (though they're all fairly close aside from SCH which has some optimization stuff going on with Energy Drain and Dissipation), and DNC, MCH, and SMN seem to be the narrowest bands to my eyes. These are also generally considered the easiest Jobs in the game. Note that this holds in both P5S (a fight of higher skilled players which largely have the fight down and so can play their Jobs pretty optimally) and Euprhosyne (I use the 24 mans as the best picture of "the playerbase as a whole" since so many people run them and there being 24 in an instance means there are good odds A LOT of runs include at least one person running a forbidden add-on that will upload, meaning the data is probably more complete than any other activity in the game).

    So this gives us a good snapshot of both the top 30% or so of players (people who do Savage) and we can compare that against the playerbase as a whole (people who do 24 mans).

    Note that while there's a TON wider variance in 24 mans because you have "all types" throwing in on those vs Savage which is a more homogeneous sample, the general pattern seems to largely hold that the same Jobs with narrower bands in Savage (the least to optimize) tend to have narrower bands in the 24 man as well. Conversely, the ones with wide bands in one tend to have wide bands in the other. Healers are a particular oddity in 24 mans, likely due to having a wider spread of playstyles (some focusing way more on healing and some going full damage and letting their co-healer take the wheel) vs Savage Healers which are pretty much all more focused on at least doing passable damage. This is probably going on with Tanks a lot there, too.

    In any case, I think it's fair to say that the easiest Jobs to optimize are:

    WAR, SGE, DNC, MCH, and SMN.

    These Jobs are also easy to pick up (SGE is the only one that's a bit hard to get the hang of if you aren't good at Healing, but it's not "hard" per se; note that WHM is almost right there with SGE, and is also considered easy to pick up and play), meaning there's a decent amount of overlap between "easy to learn" and "easy to master" in this game. There are a few exceptions (like RDM and BLM are both easy to understand on a conceptual basis, but they both actually have a lot to optimize to play well - people MAJORLY underestimate how hard RDM is to optimize...), leading to wider bars despite them being easy to understand on a basic level and do passably with.

    You can go to their site and look at the specific numbers if you want to calculate the exact widths of the bars, but I'm just doing a cursory "eyeballing".

    .

    In any case, I don't think damage should be based on "how hard it is", since different things will be hard for different people. As I say, SGE is the easiest Healer to optimize, but that doesn't mean it's necessarily the easiest FOR YOU or anyone else to play. Many people love WHM because of it being easy, but some people genuinely find it hard to heal with because it lacks tools of the more robust kits of SCH, AST, or even SGE. GNB is supposedly the hard Tank to play, but I find it easier than the other three because the rotation is rigid enough around its CDs that I never get "lost" in the rotation, while I frequently forget about WAR's self-buff and let it fall off. Presumably that'd be an "easy" thing to someone else, but my brain just doesn't track upkeep buffs/debuffs well unless they're blindingly in my face like NIN's since it's front and center on the Job Gauge with the pinwheel of kunei.

    I think people should be allowed to play the one they enjoy without being penalized.

    And looking at the 24 man spreads, even the wide bars aren't THAT much wider than the narrow bars there. So when looking at the playerbase as a whole, not just the subsection of Savage raiders, the bars are all pretty wide and pretty close to each other in wideness, meaning most Jobs are roughly the same level of hard when it comes to the playerbase as a whole.

    That said, if you want to try a role and are looking for the lowest difficulty curve of each one:

    WAR (for Tank), WHM (Healer), probably SAM (Melee), DNC (Ranged), and SMN (Caster) are the easiest to pick up and play well and to optimize well. This WILL depend on you and your brain as a player, though!

    SGE is there for Healer if you can get used to their kit (though they do ridiculous healing for a non-"pure" Healer), MCH is there for Ranged (but requires more APM and setup than DNC), and RDM is there for Casters (though requires more to optimize at the high end then SMN and does less damage on average while offering more utility; RDM is the least mobile of all Jobs in the game, less mobile than BLM, ironically). The second-hardest Melee is PROBABLY (surprisingly to many) NIN, and the second easiest Tank looks to be DRK.

    But WAR/WHM/SAM/DNBC/SMN are probably the easiest to learn in each role.
    (1)
    Last edited by Renathras; 05-14-2023 at 02:35 AM. Reason: EDIT for length

  8. #68
    Player
    Lyth's Avatar
    Join Date
    Jul 2015
    Location
    Meracydia
    Posts
    3,881
    Character
    Lythia Norvaine
    World
    Gilgamesh
    Main Class
    Viper Lv 100
    If you're interested in objective measures, you're right in that the 'spread' of data relates to the variation in performance. If you have data charted as a boxplot, there's a really simple way to do this: the interquartile ratio.

    The interquartile ratio is the difference between the 75th and the 25th percentile of a dataset. Visually, that corresponds to the 'box' of the boxplot. The reason why this is a useful measure than the variance in this type of analysis is because it is less influenced by adding an extreme outlier to the results (i.e. an unusually good or unusually poor performance.)

    This should be taken with a grain of salt though, as you're not really measuring player 'effort'. What you are measuring is the 'reward' for that effort. So a large degree of 'spread' isn't necessarily indicating that there's a greater variation in 'skill', but rather an indication of how much more good players get rewarded over weaker players. This is also harder to assess on support roles, because you may see one healer put out more damage simply by making their co-healer solo heal the encounter at the expense of their combined dps.

    You can also look at the skewness of the data, which will give you some insight into how skill is distributed across the job. A stronger positive skew is going to indicate that that there's a smaller number of 'strong performances'.

    If you want to look at effort directly, then it becomes a lot more difficult. Do you look at how 'actions' per minute correlates with performance on each job? But then are we biased in our definition of 'difficulty'? Mechanical 'difficulty' is one thing, but what about the sort of player who plans their GCDs and movements out in detail on a spreadsheet in advance? They might have time to grab a coffee between hour-long casts without needing to ever move, but do we discriminate against that simply because they aren't doubleweaving on every other GCD? Is there 'difficulty' in finding creative solutions to uptime that only a small subset of the playerbase can figure out (i.e. gamesense vs. mechanics)? It all depends on player perspective. Different players have different 'experiences' of difficulty, which is why it's so hard to assess objectively.

    Lastly, it's worth noting that while the playerbase has defined a variety of variables to describe performance, there's a degree of arbitrariness to the variables that they come up with (aDPS and nDPS are examples of this, regardless of what you think of their value in analysis). I think if you were seriously interested in studying how job performance data relates to 'difficulty', it would make for a straightforward machine learning problem, simply because you could study how different permutations of both currently derived and as yet underived variables correlate with and influence usage metrics like 'player clears'. I wouldn't be surprised if that's the way the battle team tackles it, but who knows.
    (2)

  9. #69
    Player
    Renathras's Avatar
    Join Date
    Dec 2014
    Posts
    2,747
    Character
    Ren Thras
    World
    Famfrit
    Main Class
    White Mage Lv 100
    Quote Originally Posted by Lyth View Post
    ...
    True, and that's what I meant by no metric being perfect. Why I think the spread is a decent (not flawless) metric, though, is by comparing it with subjective perspectives.

    Subjectivity is hard to measure or quantify, and obviously subjective...but that doesn't mean it has no meaning. For instance, if essentially 90+% (or a nebulous "the vast majority") feel that SMN is easy, then SMN is PROBABLY easy. This is useful in testing metrics. If there was a proposed objective metric that showed SMN as the hardest Job in the game, we could safely say that metric is measuring something, but not difficulty. On the other hand, if we look at an objective, quantifiable metric that DOES show SMN to be the easiest, and this agrees with the subjective community perspective, we can use that as a point in favor of that metric's validity, even if it's imperfect. We just have to recognize it's not THE WHOLE story.

    That was why I pointed out looking at the variance within roles (instead of across all Jobs), for example, specifically pointing out Healers having some other considerations.

    I think the fact that the variation on the forbidden sight does have narrow bars for the Jobs considered easier (WAR, SMN, DNC) and wider bars for the ones considered harder (BLM, DRG, SCH), there's probably some accuracy to it. I don't remember the term in statistics (I think it's "R-squared value"), but the one that means "this model explains ~x% of the observed behavior"; I think the variation probably should be assigned something like 50-70%. There are other things that obviously come into play, like APM (higher APM will generally be harder than lower APM; though note APM doesn't count target swapping, so AST might be more difficult than APM would suggestion, and it doesn't factor in strategic fight knowledge, so BLM appears "easy" by that metric since it has a low APM due to long cast times and few weaves), technical complexity, and things that can be hard to some people but easy for others (some people are great are handling procs or juggling buffs/DoTs while others suck at it; some are great at handling a rigid rotation while others suck at that), so ultimately, any given metric can only be so good and comes with those caveats.

    As I say, many people find WAR the easiest Tank and GNB among the hardest, but I find GNB easiest for me to play well since the rotation is more rigid and I don't have to think about juggling upkeep buffs, and GNB's powder charge is a much simpler system to my brain than remembering where in fights to pool WAR Rage; all I need with GNB is "have 3 charges going into No Mercy", which my brain tracks a lot better than "upkeep buff and pool X amount of Rage for Inner Release", even though the two SHOULD be more or less identical on paper. Somehow "3" is easier for my brain than "100 +2x 50 + X that next Weaponskill will give". So what individuals find easier won't NECESSARILY track the metric, even all other things being equal.

    ...it's also why I don't think damage should be based on "difficulty", since that IS subjective.
    (1)
    Last edited by Renathras; 05-15-2023 at 04:17 AM. Reason: EDIT for length

  10. #70
    Player
    Basteala's Avatar
    Join Date
    Apr 2022
    Posts
    400
    Character
    Basteala Thayne
    World
    Mateus
    Main Class
    Gladiator Lv 90
    Out of the DPS I've played, easiest to hardest. I'm going to take this from a relatively new player perspective, or at least new to DPS, as I think that's when ease of play is going to be most impactful:

    SMN -- You have a really simply rotation, you have very few casts, very few OGCDs, and honestly it feels really hard to screw things up, and I don't think it's too punishing even if you do.

    MCH -- I never really understood the difficulty here. It's definitely got a busy burst window, but even some tanks can say that. Of course if you have poor ping you're SoL--weaving with short GCDs on bad ping is NOT going to happen.

    RDM -- The rotation is honestly pretty easy. Your problem here is going to be more a matter of micro management, and preparing your fairly jank mobility ahead of time. You have dualcast, you have a gap closer and extender sure, but with the way modern FF14 likes to make players dance, none of it is really helpful. The OGCD drift can throw people off as well if they don't know how to handle it.

    SAM -- You got a busy burst window, and unlike the above classes, you have to worry about positionals. Although positionals for SAM are really on your terms since you have three core rotations you can put in any order. The castbar might catch you off guard. Slide casting makes it easier, even if it's a little weird to be doing that as melee, lol.

    DRG -- There are...a lot of buttons here. I think the second hardest thing about DRG is just how many buttons you have to press. Your ergonomics for your hotbars WILL be tested. There are a lot of positionals you need to keep in mind, and a lot of OCGs, as well as a couple secondary resources. The animation lock from jumps can be a pain--or your death, if you're careless.

    BLM -- Black Mage is kind of funny because the rotation really isn't all that complicated--you do your dot uptime, you do fire, then ice, rinse, repeat. The devil's in the details and in the execution, working around long cast times in an AoE heavy expansion like Endwalker has its challenges, and if you tunnel vision your dodging and don't keep after your fire/ice uptime, you will lose your fire/ice state and that can be really problematic for your damage. As others said, you really need to know the fight, as it strongly influences when you want to use leylines, triple cast, etc. Obviously you want to keep your abilities going on cooldown, but BLM has this situation where if you mindlessly put them on CD and then have to move a lot, you wind up wasting it. I don't think it's by any means bad, you have all the tools to tackle your problems, but getting even adequate with Black Mage is a matter of dedication. Same for Dragoon, honestly.
    (0)

Page 7 of 7 FirstFirst ... 5 6 7