Open benleen opened 3 years ago
thank you!
------ Original Message ------ From: "Ben Leën" notifications@github.com To: "Tercioo/Details-Damage-Meter" Details-Damage-Meter@noreply.github.com Cc: "Subscribed" subscribed@noreply.github.com Sent: 08-Mar-21 2:31:21 PM Subject: [Tercioo/Details-Damage-Meter] Feature Request: Time Weighted Damage Meters (#199)
Problem statement End of dungeon "Damage Done" or "Average DPS" is not a great indicator of a player's contribution to the timely completion of an M+ dungeon. The community however is obsessed with “average DPS” and is all too often using this as an invalid projection of their class damage effectiveness to complete the dungeon in time. Hence I suggest “Damage Done” scored against intervals of time of eg. 10seconds and relative to the best performer in the interval.
Example A 2 player party doing a dungeon:
PullEncounter:Player 1 DPSPlayer 2 DPS 1 A boss fight lasting 60 s 6k 4k 2 A trash pack fight lastering 60s 8k 12k Dungeon Average 7k 8k At first glance it looks as if player 2 did better and "contributed most" to the timely completion of the dungeon given the average dps. This is however not true as player 1 saved as much time on pull 1 compared to how much player 2 saved on pull 2 relative to one another.
Proposal Given 10 second intervals:
PullEncounter:Player 1 DPS Efficiency relative to top performer in intervalPlayer 2 DPS Efficiency relative to top performer in interval 1 A boss fight lasting 60 s 6 100% 6 66% 2 A trash pack fight lastering 60s 6 66% 6 100% Dungeon Average 83% ( (600+396)/12 ) 83% ( (396+900)/12 ) Relative player to party DPS Efficiency:
Player 1: 50% (7k AVG DPS)Player 2: 50% (8k AVG DPS)Conclusion Despite still being rudimentary, it is still a vast improvement over "average dps" to illustrate class damage effectiveness in an individual dungeon run.
I hope this is a feature you can add to Details to improve how class meta is (wrongly) perceived as well as help people improve themselves.
Note If you have no interest in building this feature I would greatly appreciate any help you can give to implement the feature myself (natively or as a plugin).
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Tercioo/Details-Damage-Meter/issues/199, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACEVKWH276BJY63W56EKFDTTCUCWTANCNFSM4YZZ3DRQ.
https://docs.google.com/spreadsheets/d/1ebZQDpgmA0SGktYYvNbIBbCKdKKOadd2cY0fWeC5iJU/edit#gid=0
With some further breakdowns/examples.
As I mentioned. I'm happy to contribute myself if you're willing to walk me through the code some time.
This would not be time-weighted dps. It would be time-and-contribution-weighted dps. The current metric is more accurate. In the example that you gave more damage had to be done during the trash pulls. So doing a higher fraction of the trash damage did contribute more to the timeliness of the dungeon than doing a higher fraction of the boss damage. And that's accurately reflected by the current metric.
Problem statement
End of dungeon "Damage Done" or "Average DPS" is not a great indicator of a player's contribution to the timely completion of an M+ dungeon. The community however is obsessed with “average DPS” and is all too often using this as an invalid projection of their class damage effectiveness to complete the dungeon in time. Hence I suggest “Damage Done” scored against intervals of time of eg. 10seconds and relative to the best performer in the interval.
Example
At first glance it looks as if player 2 did better and "contributed most" to the timely completion of the dungeon given the average dps. This is however not true as player 1 saved as much time on pull 1 compared to how much player 2 saved on pull 2 relative to one another.
Proposal
Relative player to party DPS Efficiency:
Conclusion
Despite being rudimentary, it is still (IMO) a vast improvement over "average dps" to illustrate class damage effectiveness in an individual dungeon run.
I hope this is a feature you can add to Details to improve how class meta is (wrongly) perceived as well as help people improve themselves.
Note
If you have no interest in building this feature I would greatly appreciate any help you can give to implement the feature myself (natively or as a plugin).