Guest
home
beatmaps
rankings
community
store
help
forum
  1. Forums
  2. osu!
  3. Development
  4. Feature Requests
  5. Completed Requests

Web: Efficiency definition [Waiting]

posted
Total Posts
12
This is a feature request. Feature requests can be voted up by supporters.
Current Priority: +0
eyup
osu! Alumni
549 posts
Here since the beginning
Topic Starter
eyup
At the moment, Efficiency = Ranked Score / Total Score. I don't think this ratio is meaningful at all? How about Efficiency = (Number of S/SS/SH/SSH) / Total Number of Beatmaps Played?

For instance, Saturos would have a jawdropping efficiency of 97.87% (506 out of 517 maps are either S or SS!) whereas peppy has an efficiency of 16.1 I MEAN 100%.

Maybe that will encourage players to shoot for more S ranks? Maybe also have a Hidden Efficiency for those who like their beatmaps served with nothingness?
peppy
19,301 posts
Here since the beginning
peppy
And eyup comes in for the killing.

Ouch
eyup
osu! Alumni
549 posts
Here since the beginning
Topic Starter
eyup
This map has been deleted on the request of its creator. It is no longer available.
Gemi
osu! Alumni
685 posts
Joined March 2008
Gemi
Agreed with the original suggestion, but maybe count S/SH as 0.5 and SS/SSH as 1 for the calculation?
ROMaster2
216 posts
Joined February 2008
ROMaster2

Gemi wrote:

Agreed with the original suggestion, but maybe count S/SH as 0.5 and SS/SSH as 1 for the calculation?
That'll make me rocket higher than necessary.
adamskii_uk
324 posts
Joined April 2008
adamskii_uk

eyup wrote:

At the moment, Efficiency = Ranked Score / Total Score. I don't think this ratio is meaningful at all? How about Efficiency = (Number of S/SS/SH/SSH) / Total Number of Beatmaps Played?

For instance, Saturos would have a jawdropping efficiency of 97.87% (506 out of 517 maps are either S or SS!) whereas peppy has an efficiency of 16.1 I MEAN 100%.

Maybe that will encourage players to shoot for more S ranks? Maybe also have a Hidden Efficiency for those who like their beatmaps served with nothingness?
Your idea isn't going to work because if you have two players, one with 200/200 'S' rank scores and the other with 200/200 'SS' rank scores, they are both going to get 100% efficiency even though the person who got the 200 'SS' rank scores did a lot better since he got 100% accuracy on every beatmap he played. Your idea is still good though, it just requires some additional tweaking to get it to work. :)

IMO, I think the efficiency score should be calculated to include the difficulty level of a beatmap that was played (1...5 stars), since getting a 'SS' rank on a 1 star beatmap is no way near as impressive as getting a 'SS' rank on a 5 star beatmap, such as getting a 'SS' rank on Shotgun Symphony [Insane].
eyup
osu! Alumni
549 posts
Here since the beginning
Topic Starter
eyup
Good points, adam, but I think the most important thing about this stat (apart from relevancy, which this suggested change is meant to address) is its ease of interpretation. No point looking at the stats table and thinking "50% efficiency, which means... ahh... he has SS'ed a weighted average of... err... 3 star beatmaps averaged but minus Hidden mods... umm fuck it I'm going to look up pr0n"

The way I've suggested means you can look and straight away know that this player has S'ed or SS'ed 68% of his/her beatmaps played. Perhaps if you really wanted to see who's the "better" player then it might pay to introduce S efficiency and SS efficiency but that's a bit of overkill IMO.
awp
osu! Alumni
11,869 posts
Here since the beginning
awp

eyup wrote:

Good points, adam, but I think the most important thing about this stat (apart from relevancy, which this suggested change is meant to address) is its ease of interpretation. No point looking at the stats table and thinking "50% efficiency, which means... ahh... he has SS'ed a weighted average of... err... 3 star beatmaps averaged but minus Hidden mods... umm fuck it I'm going to look up pr0n"

The way I've suggested means you can look and straight away know that this player has S'ed or SS'ed 68% of his/her beatmaps played. Perhaps if you really wanted to see who's the "better" player then it might pay to introduce S efficiency and SS efficiency but that's a bit of overkill IMO.
I don't think there's really a large enough difference between S and SS to merit aggregate distinction. Also, eyup, as to what you were saying, I'm not sure I really

ah fuck it I'm going to go watch porn
adamskii_uk
324 posts
Joined April 2008
adamskii_uk
As I've mentioned already in my previous post, there is still a MASSIVE difference between getting a 'S' or 'SS' rank on 1 star beatmap, and getting the same rank on a 5 star beatmap. Unfortunately, at the moment there's no way to tell for sure how many 'S' and 'SS' ranks people have accumulated on the easier beatmaps. When I see someone who has 300+ 'S' ranks on their stats, over half of them could be from very easy beatmaps, at their normal speed, with no MOD's added, etc.

If it's easy to do, I think the 'Ranking Distribution' page for the stats would be a lot better if it showed how many 'S' and 'SS' ranks were gotten for each beatmap level, rather than just lumping it all together in one group. Then we can see exactly how many 'S' and 'SS' ranks people have accumulated on the easy beatmaps, and how many a person has got on the difficult beatmaps.

awp wrote:

I don't think there's really a large enough difference between S and SS to merit aggregate distinction.
I guess so. However, I would still rather get a 'SS' rank though. ;)
Last edited by adamskii_uk , edited 1 time in total.
Saturos
osu! Hall of Famer
1,630 posts
Here since the beginning
Saturos
In that case, it could be shown in a lump total on the ranking page as per usual, and then broken down within the newly sexified profile page. Though I doubt it'd be able to split apart modded S's.
Gemi
osu! Alumni
685 posts
Joined March 2008
Gemi

adamskii_uk wrote:

If it's easy to do, I think the 'Ranking Distribution' page for the stats would be a lot better if it showed how many 'S' and 'SS' ranks were gotten for each beatmap level, rather than just lumping it all together in one group. Then we can see exactly how many 'S' and 'SS' ranks people have accumulated on the easy beatmaps, and how many a person has got on the difficult beatmaps.
It's been planned to show the ranks per difficulty level just like the accuracy tab shows accuracy per difficulty level.
Gemi
osu! Alumni
685 posts
Joined March 2008
Gemi
During the community meeting it was decided that Efficiency will be changed to:

Efficiency = Number of passed songs / Number of plays

Ie every time you play a song and pass it both Number of passed songs and Number of plays get +1. Every time you play and don't pass only Number of plays gets +1.

Please sign in to reply.

New reply

0
1 / 12