1. #48781
    The Insane Granyala's Avatar
    Join Date
    Feb 2010
    Location
    Arkon-III
    Posts
    19,756
    Quote Originally Posted by KrayZ33 View Post
    How would *you* read this, knowing I was playing with the same people and classes (AST, RDM, DNC, BLM - GNB, WAR, WHM and ME(SAM))? what would you read if you only had the green box?
    I would not read it at all.
    DPS number is only a quick indicator to see whether I ended up in the "to be expected" range or whether there might be sth up that I missed during combat.
    Any actual analysis and improvements come from looking far deeper into the log.

    Try 1 vs 2 have a delta of 230 aDPS. That may be crit luck / having to move at a more inopportune moment etc.
    TBH: I wouldn't even bother researching that tiny of a difference. Not worth my time.

  2. #48782
    Quote Originally Posted by Granyala View Post
    I would not read it at all.
    DPS number is only a quick indicator to see whether I ended up in the "to be expected" range or whether there might be sth up that I missed during combat.
    Any actual analysis and improvements come from looking far deeper into the log.

    Try 1 vs 2 have a delta of 230 aDPS. That may be crit luck / having to move at a more inopportune moment etc.
    TBH: I wouldn't even bother researching that tiny of a difference. Not worth my time.
    Exactly, and when you glance over at #3 and would only look at parsed DPS, you'd think they are equal too.
    But aDPS and rDPS show that #3 is much worse than #2 and #1.

    As mentioned a few times by now. Neither number will tell you "I have to delay XY by 3 seconds to get the buff in".
    But you can see at a glance that something is wrong that is worth looking for if you are interested in finding out what was wrong.

  3. #48783
    Quote Originally Posted by Hyral View Post
    From others posts and threads I've seen you on, we are of one mind about this game. So I'm not surprised your idea sounds really fun to me and nobody else lol. I'm not sure how parsing works, never really bothered with it, but by what little I can infer from other people talking about it what you are doing is kind of like those "ghost trials" in racing games where you compare your own performance with your previous best and try to beat it. I might start looking into this if that's what it is, really sounds like something I could sink some time into. Especially once I get all my jobs to 80.
    You're not alone! Leveling with spotify on is really fun, and I just wish I could parse myself and see my numbers, but I play on Ps4, so no tools for me

  4. #48784
    The more I look at this screenshot they gave us, the more hints I see. (or maybe I'm grasping at straws? No spoilers, just my own speculation)



    If you count the ancients, it makes 13. The hue of the image suggests the WoL is reliving a memory via the echo. We know the 14th abandoned the group over a difference of opinion on how to deal with the calamity. Could it be we're reliving the point of view of the 14th being stared down by the other 13 convocation members? It was hinted in 5.0 that the WoL was the 14th member of the convocation that left the group, so it could be our own memory of a previous life. But still, knowing Yoshi, it could just as well be the memory of someone else we're experiencing. Like Elidibus. Could Elidibus be the 14th instead? (The new/current Elidibus, not the predecessor that was the catalyst for Zodiark) It's certainly curious when you consider how different Elidibus is from all the other Ascians. He plainly tells Varis that if he wants to kill his fellow dark robed ascians, he wouldn't stand in his way. Makes you wonder.

  5. #48785
    The Insane Granyala's Avatar
    Join Date
    Feb 2010
    Location
    Arkon-III
    Posts
    19,756
    Quote Originally Posted by Merie View Post
    It was hinted in 5.0 that the WoL was the 14th member of the convocation that left the group, so it could be our own memory of a previous life
    We're the partially reconstructed 14th member, yes.
    That much has been clear since 5.0 when a certain someone recognizes us.

    However, we do not have the memories of the 14th, so we don't need to worry about switching to the dark side.
    Remember though, Echo does not show our own memory but s/o else's.

  6. #48786
    Quote Originally Posted by Granyala View Post
    Remember though, Echo does not show our own memory but s/o else's.
    We could be reliving the memories of one of our other incarnations. Perhaps Unukalhai.

  7. #48787
    Quote Originally Posted by Granyala View Post
    We're the partially reconstructed 14th member, yes.
    That much has been clear since 5.0 when a certain someone recognizes us.

    However, we do not have the memories of the 14th, so we don't need to worry about switching to the dark side.
    Remember though, Echo does not show our own memory but s/o else's.
    It was hinted, but was it definitively confirmed? Hythlodaeus spoke of the 14th leaving the group and also spoke of knowing us and us having a relationship with hades. But while being the 14th would check that box, I think it could be just as possible that that was just Yoshi's usual misdirection to subvert our expectations to surprise us later. It's possible Emet Selch remembers us not for being a peer on the council, but for being a son/daughter instead.

    And yes the echo shows someone else's memory. So maybe we're looking through the eyes of the 14th in the screenshot but not necessarily our own.

  8. #48788
    Quote Originally Posted by KrayZ33 View Post
    You are currently trying to tell me that 3 sets of information < 1 set of information (which is also included in the 3 sets from before)
    No, I'm trying to argue that using rDPS as the go-to standard measurement when it comes to rankings is a dumb idea. The fact that you've produced logs where you've done less total DPS, but higher rDPS only helps prove that point.

    Additionally, I've been arguing that it's useless as a self-comparative metric because it doesn't take into account how well you've coordinated and worked with your team. It also takes into account other peoples performance into your own if you're a job that can bring a group buff. It encourages blindly performing your rotation without considering how you're going to optimise, and that's a dangerous precedent to set.

    The metric that really matters when it comes to determining results is just the straight forward DPS one. I don't think it's coincidental that your fastest kill is also the one where you did the most DPS - Even if it's not the one where you did the most rDPS. The kill you did the most rDPS is the slowest kill. Which, for me, is the biggest flaw with rDPS of all. It correlates with your DPS, but isn't always causal. It's useful to know, but should not be the de-facto default measurement used on FFLogs when it comes to rankings.

  9. #48789
    Quote Originally Posted by StrawberryZebra View Post
    No, I'm trying to argue that using rDPS as the go-to standard measurement when it comes to rankings is a dumb idea. The fact that you've produced logs where you've done less total DPS, but higher rDPS only helps prove that point.
    It doesn't help at all, because, as mentioned, #2 is a run with a dancer and #1 is a run with the same dancer but he got a different partner

    Additionally, I've been arguing that it's useless as a self-comparative metric because it doesn't take into account how well you've coordinated and worked with your team. It also takes into account other peoples performance into your own if you're a job that can bring a group buff. It encourages blindly performing your rotation without considering how you're going to optimise, and that's a dangerous precedent to set.
    why don't you use aDPS then - which is also used as ranking metric, that will remove single-target buff-padding such as dancer buffs, but will show how well you used group buffs.

    The metric that really matters when it comes to determining results is just the straight forward DPS one. I don't think it's coincidental that your fastest kill is also the one where you did the most DPS - Even if it's not the one where you did the most rDPS. The kill you did the most rDPS is the slowest kill. Which, for me, is the biggest flaw with rDPS of all. It correlates with your DPS, but isn't always causal. It's useful to know, but should not be the de-facto default measurement used on FFLogs when it comes to rankings.
    No.
    It doesn't, like - at all.
    I mean, you are basically doing the mistake I was talking about. You think the try with the most "green"DPS was my best one.
    When it's far more likely (and that's how it actually is) that the performance of the other players was a lot better.
    There is de-facto no benefit of having the "green"DPS-metric as the main metric because it does jack shit about telling you how "well" you used group buffs because it doesn't even show how much ST buffs you stacked, and judging from your post, will lead to false assumptions. So if it's not even better in that regard, why even use it?

    And *not* pressing cooldowns mindlessly, actually means that Dancers/Classes with buffs (which is the majority), get a better result in rDPS and thus a better ranking. So smart usage (stacking and full time usage) of cooldowns is actually incentivised, not the other way around.

    rDPS is group and support oriented, aDPS is slightly support oriented and slightly padding - and DPS is for padding.
    You are only looking at things from a SAM/BLM perspective - or Tank perspective.
    For classes that actually deliver support, "green"DPS is nearly meaningless.

    AST without using cards but spamming a DPS spell would get a better "green"DPS result, but would be a worse player.
    And since the Logs aren't supposed to make people stack 2 DNC with 2 DPS for padding reasons, they use rDPS
    Last edited by KrayZ33; 2020-07-31 at 06:46 AM.

  10. #48790
    The Insane Granyala's Avatar
    Join Date
    Feb 2010
    Location
    Arkon-III
    Posts
    19,756
    Quote Originally Posted by KrayZ33 View Post
    For classes that actually deliver support, "green"DPS is nearly meaningless.
    Aye, that much is true. You miss so much data about the benefit of your support to others.

    Still, if you exclude the buffs to others, you get to see how you performed, which can be obfuscated by rDPS to some degree.
    So I think there is merit in having both values.

  11. #48791
    On a completely different topic, I was doing some achievement hunting in WoW today to take a break with something light before our raid night in FF and I kept thinking to myself... Why doesnt FFXIV make some fun achievements?
    All the achievements are basically "Do dungeons 1 trillion times", "Do this raid 15 times", "kill a bajilion enemies" like this is not fun at all and not something you d wanna chase but rather randomly just get over time. Also the achievement UI thing is a mess to look at. Its gray and dull and looks very unappealing.
    I think the game could benefit a lot from having some more fun achievements instead of long drawn out achievements that you just get at some point purely from playing the game.

  12. #48792
    Quote Originally Posted by Delever View Post
    On a completely different topic, I was doing some achievement hunting in WoW today to take a break with something light before our raid night in FF and I kept thinking to myself... Why doesnt FFXIV make some fun achievements?
    All the achievements are basically "Do dungeons 1 trillion times", "Do this raid 15 times", "kill a bajilion enemies" like this is not fun at all and not something you d wanna chase but rather randomly just get over time. Also the achievement UI thing is a mess to look at. Its gray and dull and looks very unappealing.
    I think the game could benefit a lot from having some more fun achievements instead of long drawn out achievements that you just get at some point purely from playing the game.
    Yeah, I'm not sure if they don't exist at all tbh, there might be a few, but they could actually make achievements for competitive players too.
    Like "kill boss X before the second Y" - Unlock title: "..of the E-peen"
    Or just some fun ones that require you to do a mechanic in a certain way... like the Ulduar ones in WoW... gosh, these were fun!

    On the other hand... can you look up another player's achievements? I could go without "above the curve" requirements for random savage runs. But I believe players don't really look up FFlogs either when players join the group... so that might not even be an issue.
    Last edited by KrayZ33; 2020-07-31 at 12:07 PM.

  13. #48793
    Quote Originally Posted by KrayZ33 View Post
    Yeah, I'm not sure if they don't exist at all tbh, there might be a few, but they could actually make achievements for competitive players too.
    Like "kill boss X before the second Y" - Unlock title: "..of the E-peen"
    Or just some fun ones that require you to do a mechanic in a certain way... like the Ulduar ones in WoW... gosh, these were fun!

    On the other hand... can you look up another player's achievements? I could go without "above the curve" requirements for random savage runs. But I believe players don't really look up FFlogs either when players join the group... so that might not even be an issue.
    I was thinking of these with the fun ways of doing mechanics like in WoW raids but sadly with the way that FFXIV encounters currently work I dont think this is possible.
    They would have to fundamentally alter fight design to be more reactive and less scripted and do away with the "oops mistake, insta raid wipe" mechanics.
    Currently most mechanics in FFXIV fights have a very specific way in which they should be executed and if you fuck up its pretty much wipe. The only thing that often changes is that people often find different positionings that work for some mechanics but the resolution of the mechanical pattern ends up being the same.
    I remember the bouncy ball achievement you could do in dragon soul which was super fun. It make the fight a ton more challenging but its not an insta wipe if you deviate from the mechanics if you can do it in a manageable way. This would have to change in FF in order for such achievements to exist.

  14. #48794
    Quote Originally Posted by granyala View Post
    damn.. Why do i still automatically read that in her voice 11 years later?!

    I've been damaged. .__.
    bbbbbbbbbbbbbbeeeeeeeeeeeetrrrrrrrrrrrrrraaaysss youuuuuu

    *******************

    you know it surprises me how many facial expressions you can pull off on your character in FFXIV

    I wish you could choose expressions to be permenant.
    Last edited by Tenjen; 2020-07-31 at 10:43 PM.

  15. #48795
    The Insane Granyala's Avatar
    Join Date
    Feb 2010
    Location
    Arkon-III
    Posts
    19,756
    Quote Originally Posted by Tenjen View Post
    I wish you could choose expressions to be permenant.
    Yes I wished that for a long time. My cat would basically heal with her eyes closed.
    Best you can do is make a macro with 15 lines, giving you 75 seconds.

  16. #48796
    On every skill? That could work.

  17. #48797
    All the aDPS/rDPS B's and I'm over here like "let's just go for fastest fuckin' kill speed"

  18. #48798
    The Insane Granyala's Avatar
    Join Date
    Feb 2010
    Location
    Arkon-III
    Posts
    19,756
    Quote Originally Posted by Kazela View Post
    "let's just go for fastest fuckin' kill speed"
    Which is basically DPS and nothing else.

  19. #48799
    Quote Originally Posted by KrayZ33 View Post
    You think the try with the most "green"DPS was my best one.
    Because, statistically speaking, it is your best one. You gained 676 DPS from your group buffs on it, while on your next closest one was 524.9 DPS. That's close to a 30% difference. You also gained 1909.4 DPS from your single target buffs, compared with your next best of 1773 - Almost 10% extra, in otherwords. These are statistically significant no matter how you look at them.

    You're right in that it might not be all to do with you, however your group performed better and it shows. You should be doing whatever it was you did differently going forwards.

    Quote Originally Posted by KrayZ33 View Post
    why don't you use aDPS then - which is also used as ranking metric, that will remove single-target buff-padding such as dancer buffs, but will show how well you used group buffs.
    That would not be a better one to use no. FF14 logs uses rDPS as it's primary ranking metric. Which is dumb because it looks at one player in a vacuum, not as part of a wider group performance. rDPS alone is a very poor metric because it excludes the other data needed to give it context, aDPS does too, where regular DPS does not.

    For you as a Samurai, rDPS is you railing against a training dummy. That's fine when it's just yourself, but it's only loosely correlated to your actual DPS output in a group setting.

    All of this is simply to stop people whinging about Dancers padding parses. Which is also an irrational line to take, the entire group role of the Dancer is to do just that. If I were a betting man, I'd be prepared to put money on Squenix doing it intentionally to appeal to that particular audience. People have been meter padding since as long as I remember there being Damage meters. Squenix seem to have embraced the idea that players are going to do it, it's time for the players to do the same in my opinion.

    Quote Originally Posted by Tenjen View Post
    I wish you could choose expressions to be permenant.
    My character pretty much wears a permenant scowl anyway. I don't see any reason why I'd want to change that.

  20. #48800
    Quote Originally Posted by StrawberryZebra View Post
    Because, statistically speaking, it is your best one. You gained 676 DPS from your group buffs on it, while on your next closest one was 524.9 DPS. That's close to a 30% difference. You also gained 1909.4 DPS from your single target buffs, compared with your next best of 1773 - Almost 10% extra, in otherwords. These are statistically significant no matter how you look at them.
    statistically speaking, #1 was my best, because I did more aDPS. Which is my own DPS + Raid buff DPS.
    So no matter how "well" or "bad" I was at using raid buffs for my own DPS, in the end, I did more DPS by either lining, or not lining up my high hitters with raid buffs.

    For you as a Samurai, rDPS is you railing against a training dummy. That's fine when it's just yourself, but it's only loosely correlated to your actual DPS output in a group setting.
    again - use aDPS then, which has it's own ranking, it's not a "secondary" ranking. It's the "other" ranking. You can sort your rankings by either of the two.

    You keep repeating your stuff, but in the end, you aren't getting a single benefit from just looking at "green"DPS.
    It's silly, it really is. You are telling me how xDPS is dependend on Y, but that is the truth for every single metric shown there.

    That would not be a better one to use no. FF14 logs uses rDPS as it's primary ranking metric. Which is dumb because it looks at one player in a vacuum, not as part of a wider group performance
    exactly the same with greenDPS lol.
    You look at 1 player being padded into the top spots without looking at the other players in your raid.
    Do you even understand that your suggestsion wouldn't change anything? You are just left with a worse situation because all of a sudden, you can't choose to filter out stuff anymore, neither can Dancer or Supports see how much they contribute to raidDPS.
    Last edited by KrayZ33; 2020-08-02 at 05:14 PM.

Posting Permissions

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