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

New PP system. [Updated]

posted
Total Posts
36
This is a feature request. Feature requests can be voted up by supporters.
Current Priority: +19
-[Husky]-
7 posts
Joined November 2014
Topic Starter
-[Husky]-
Hey guys :D
We all have same probem. We first start to get easy PPs then harder and harder to get some and usullay u get 1 or 2. I hope then dev. team will notice this and use it in future :)
Osu! PP system
So, most of time I spent playing Osu! and i need hours to get 10 or 20 PPs. Now the system of giving them should work by next two formulas.
Amount of PPs that you will get

Now.
KM means coefficient of mod
Now this can improove your chance of getting some PPs.
Mods Hard Rock and Hidden (HR, HD) would have Coefficient 2
Mods Flashlight, Double Time/Nightcore (FL, DT/NC) would have coefficient 6
Now if player uses more mods they are calculated into one number.
But if player does not uses any mods its counted like "1"
Example
I play map with DT, HD and HR, so my coefficient of mod would be 2+2+6 which would be 10.
Now how this formula works and special thanks to Orion and Baraatje123 for sharing ideas and suggestions with me:
I will take ExGon on map Big Black



Conclusion:
According to this formula he/she should get 600pps.
You will probablly see that this formula follow your measurment of performance, it's simmilar to weightage system but it's not. Like this everyone will have chance to get awarded for their hard work :D
Tell me what you think about this! :D

Change log
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Removed formula number 1 becuase of ilogical sence
Last edited by -[Husky]- , edited 3 times in total.
Bara-
11,978 posts
Joined April 2013
Bara-
You don't know how pp works do you?
It's weighted, get better plays, get more pp
Bad plays shouldn't get pp, only the outstanding ones
Also, this system means spamming maps will keep giving pp, which I don't agree with
Karuta-_old_1
947 posts
Joined November 2009
Karuta-_old_1
my only problem is that peppy is trying to turn this game into a visual novel, I don't really look think pp is all that important but as a guideline

it is cool that you can make formulas but I don't agree with awarding pp due to luck and stuff

and mods are really something you should play well with not just pass
Orion
36 posts
Joined May 2014
Orion
rrtyui's Big Black 100% by this formula:
{|[(1337*1)-1500]|:[(100*100)/10]}*5
{|(1337-1500)|:[(10000)/10]}*5
(165:1000)*5
0.165*5
.825
appx a %1 chance of obtaining pp from Big Black with an SSFC

Also why are we calculating %chance and not pp value?
It shouldn't be based on chance whether you get pp from a map...
Last edited by Orion , edited 3 times in total.
worst fl player
worst fl player
#FuckPP
Bara-
11,978 posts
Joined April 2013
Bara-
Or try this
A normal, max combo is 140, AR=4
80% no mod
|(140*1)-1500|:80*100/10 *5=
|-1360|:(8000/10) *5=
1360/800 *5= 8.5%

Now with HD DT
same acc
|140*(6+2)-1500|: 800 *5 (800= 80*100/10)
|1120-1500|: 800 *5
|-380|: 800 *5
380:800*5= 2.375%

So using HD DT lowers chance of pp?
Just no
abraker
Global Moderator
8,327 posts
Joined July 2014
abraker
Why CHANCE of earning pp? This is nonsense. PP should be earned as a measurement of your performance not as a chance of gaining performance due to the measurement of your performance.

While the statement "Player X has p% chance of gaining Y pp" holds true due to uncertainties within the player's potential, just leaving it to plain luck is just a degradation of what performance comparison ought to be.
Endaris
5,282 posts
Joined June 2010
Endaris
This belongs in Gameplay and Rankings 8-)

@BlackWiddow : PLAY MORE!
There is a very good reason why you don't get pp for slapping a ton of mods on easy maps and is that an easy map with mods is still easy in many aspects when comparing it to a hard map.

For more information about pp, look here:
https://osu.ppy.sh/wiki/FAQ#Performance_Points
Minhtam
550 posts
Joined September 2014
Minhtam

Endaris wrote:

@BlackWiddow : PLAY MORE!
There is a very good reason why you don't get pp for slapping a ton of mods on easy maps and is that an easy map with mods is still easy in many aspects when comparing it to a hard map.
This is true. The majority of [Normal] maps on +HR, +HD, +DT are significantly easier than [Hard] no-mod maps.
Remyria
361 posts
Joined June 2012
Remyria

BlackWiddow wrote:

SPOILER
Hey guys :D
We all have same probem. We first start to get easy PPs then harder and harder to get some and usullay u get 1 or 2. I hope then dev. team will notice this and use it in future :)
Osu! PP system
So, most of time I spent playing Osu! and i need hours to get 10 or 20 PPs. Now the system of givin them should work by next formula.



Now.
HCP means Highest Combo Possible
KM means Coefficient of Mod
Now this can improove your chance of getting some PPs.
Mods Hard Rock and Hidden (HR, HD) would have Coefficient 2
Mods Flashlight, Double Time/Nightcore (FL, DT/NC) would have coefficient 6
Now if player uses more mods they are calculated into one number.
But if player does not uses any mods its counted like "1"
Example
I play map with DT, HD and HR, so my coefficient of mod would be 2+2+6 which would be 10.
Ⅰsome numberⅠ When u see this brackets it means that if number is negative (Ⅰ-500Ⅰ) it becomes positive (500)
KD means Coefficient of difficulty
Coefficient of difficulty depends on one factor - Approach Rate
When approach rate goes from 1 to 3 KD will be 200
Those maps are very slow and easy to pass so there will be smallest chance possible to get some PPs.
When approach rate goes from 4 to 6 KD will be 1500
Those maps are easy to pass and have decent chance og getting PPs.
When approach rate goes from 7 to 9 KD will be 1000
Now this maps are harder to pass and requires some effort there is good chance to get PPs.
When approach rate is 10 KD will be 1500
This maps are EXTREMELY hard to pass and there is BIGGEST chance of getting PPs
PA means Player's Accuracy
Now how this formula works:
Map's Combo: 2000
Mod's Using: DT, HD
Accuracy: 85%
Approach Rate: 9

Tell me what u think about this :D
Part 2 coming soon!
It's not Final Fantasy, it's Osu!...We're not calculating chances of getting a rare item, we're calculating how good a player is based on his capability of doing good scores with a little bonus on farming(the 400 and something thingy). This feature request is out of place in my opinion.

Plus, there's a detail that makes it even more go into nonsense=> Scores stay, if you do not PP for the score, you will need to BEAT IT to have another chance of getting PP's
Musty
2,252 posts
Joined February 2010
Musty
this is not pp this is score (i mean farming)
Remyria
361 posts
Joined June 2012
Remyria

Musty wrote:

this is not pp this is score (i mean farming)
But rank being set on your amount of rare artefacts you can randomly get by playing :^)
Houtarou Oreki
303 posts
Joined March 2014
Houtarou Oreki
Nah, the ppv2 is much better than this imo. The weighting of scores and everything overall is just the best thing there could possibly be, no need to change it.
-[Husky]-
7 posts
Joined November 2014
Topic Starter
-[Husky]-

Baraatje123 wrote:

You don't know how pp works do you?
It's weighted, get better plays, get more pp
Bad plays shouldn't get pp, only the outstanding ones
Also, this system means spamming maps will keep giving pp, which I don't agree with
I do but I'm sick of playing for like 2-3 hours and getting nothing. Also when i pass map with higher AR and so on i barelly get one PP

Karuta- wrote:

stuff
my only problem is that peppy is trying to turn this game into a visual novel, I don't really look think pp is all that important but as a guideline

it is cool that you can make formulas but I don't agree with awarding pp due to luck and stuff

and mods are really something you should play well with not just pass
Well with this system I want to push player's limits and yet to be rewarded :)

Orion wrote:

stuff
rrtyui's Big Black 100% by this formula:
{|[(1337*1)-1500]|:[(100*100)/10]}*5
{|(1337-1500)|:[(10000)/10]}*5
(165:1000)*5
0.165*5
.825
appx a %1 chance of obtaining pp from Big Black with an SSFC

Also why are we calculating %chance and not pp value?
It shouldn't be based on chance whether you get pp from a map...
I fixed everything and founded out where was mistake in formula and added one more, check it out and tell me what you think :D

Baraatje123 wrote:

stuff
Or try this
A normal, max combo is 140, AR=4
80% no mod
|(140*1)-1500|:80*100/10 *5=
|-1360|:(8000/10) *5=
1360/800 *5= 8.5%

Now with HD DT
same acc
|140*(6+2)-1500|: 800 *5 (800= 80*100/10)
|1120-1500|: 800 *5
|-380|: 800 *5
380:800*5= 2.375%

So using HD DT lowers chance of pp?
Just no
I've fixed formula and i would be delighted if you want to help me in further development of it :)

abraker wrote:

stuff
Why CHANCE of earning pp? This is nonsense. PP should be earned as a measurement of your performance not as a chance of gaining performance due to the measurement of your performance.

While the statement "Player X has p% chance of gaining Y pp" holds true due to uncertainties within the player's potential, just leaving it to plain luck is just a degradation of what performance comparison ought to be.
Thanks for pointing out this to me :)
I understand you... But still... I will work more on fixing forumla. This actually values effort that is put in that map, see example and I hope you will understand. If you have any suggestions in remodulating formula I would be more than happy to apply it/them.

Endaris wrote:

stuff
This belongs in Gameplay and Rankings 8-)

@BlackWiddow : PLAY MORE!
There is a very good reason why you don't get pp for slapping a ton of mods on easy maps and is that an easy map with mods is still easy in many aspects when comparing it to a hard map.

For more information about pp, look here:
https://osu.ppy.sh/wiki/FAQ#Performance_Points
I know... I'm trying hard on hard maps but I loose hope in them... On daily basics i pass like 10-15 sometimes more hard maps... I understand what you want to say but still... I think that most of our plays should be awarded...

Houtarou Oreki wrote:

Nah, the ppv2 is much better than this imo. The weighting of scores and everything overall is just the best thing there could possibly be, no need to change it.
I know but still.. I wanted to share my idea with community, I didn't killed anyone ;)
Last edited by Trosk- , edited 2 times in total.
Bara-
11,978 posts
Joined April 2013
Bara-
Please don't heptatuple post (is that a word)
Just use the edit button
Trosk-
Technical Support Team
5,795 posts
Joined October 2013
Trosk-

Baraatje123 wrote:

Please don't heptatuple post (is that a word)
Just use the edit button
^ This, please.
Bara-
11,978 posts
Joined April 2013
Bara-
How is his post count still 10
Or is that intended after merge?
Reminds me of that one person who always double posted in help >__>
abraker
Global Moderator
8,327 posts
Joined July 2014
abraker
Can you explain the reasoning and how the 2nd formula works? Why multiply everything? Why choose those constance? How did you come up with the fornula? What is the colon thing?
Maldenarus
25 posts
Joined September 2014
Maldenarus
It seems you have only basic mathematical skills. To define a formula to get an arbitrary number between 0 and 100 and then slapping an % on the back is bad. If one would like to use a percentage based system you have to come up with something that defines your upper boundry (which would be the 100%) and then calculate based upon that a difference so that one could get an 60% with the given parameters. This would have to be consistent over every parameter in that formula which i, without having it plotted, highly doubt. Having a 130% is sign for a bad system. Why having a chance over 100% in the first place. It´s illogical. The highest possible parameters put into that formula should be 100% not some number above that.

to have a chance system is apart from the very creative calculation utter crap. Why would some low AR maps give lower chance at pp than high AR maps. It would punish new players who are not able to play high AR, and it would encourage farming one map over and over until i get my pp score which isn´t the definition of pp at all.
abraker
Global Moderator
8,327 posts
Joined July 2014
abraker

Maldenarus wrote:

stuff
^^^ this


Also If you are to make a pp system from scratch, and heck, why not even throw in the perfect scoring system with it, the best way to rate performance is the following:

1) Go through beatmap data to generate the difficulty for the beatmap
  • a) Calculate the probability of completion, aka difficulty, of doing a pattern at time t. This difficulty is measured in the odds of success based on player-constants
    b) Multiply all the odds together ("and" in terms of probability)
    c) Generate max pp = C/(1-probability of completion), where C is some constant
2) Go through replay data to generate completion difficulty index
  • a) Normalized score (0.0 to 1.0) gained at point t multiplied by the odds of success at point t
3) Generate the player pp
  • a) player pp = max pp * completion difficulty index
All is left is to figure out those probability formulas and figure out the player constants, and you got yourself a pretty solid system. Sounds easy, right? I've been researching this stuff for 9 months now and trust me it's hard. Wait until I release an article on odds of hitting a hitcircle and you will see how "simple" it is.
Kurochiie
754 posts
Joined April 2014
Kurochiie

Remyria wrote:

It's not Final Fantasy, it's Osu!...We're not calculating chances of getting a rare item, we're calculating how good a player is based on his capability of doing good scores with a little bonus on farming(the 400 and something thingy). This feature request is out of place in my opinion.

Plus, there's a detail that makes it even more go into nonsense=> Scores stay, if you do not PP for the score, you will need to BEAT IT to have another chance of getting PP's
You, sir, I like you.
GhostFrog
416 posts
Joined March 2013
GhostFrog
If you want a number that goes up every time you pass a map, use level and ranked (or total) score. Performance points are meant to measure performance, not playcount.
Remyria
361 posts
Joined June 2012
Remyria
let's try my perf on NNRT with your exact equation:

((((AR*10)*accuracy)*KM/2):100)*grade

((((9*10)*99.9)*1/2):100)*5

(((90*99.9)*1/2):100)*5
((8991*1/2):100)*5
(4495.5:100)*5
44.955*5=224.775 where I get 215 with actual system

seems, decent...but there's a big problem here:

Septembre's score on Midnight Siege

((((AR*10)*accuracy)*KM/2):100)*grade

((((9.6(I don't know since your example had HR on an already AR10)*10)*83.75)*6/2):100)*3

(((96*83.75)*3):100)*3
((8040)*3):100)*3
(24120:100)*3
241,2*3

723.6 where he gets...uh...236pp...

I'll try again without applying the DT to the AR

(I skip few steps)
(((80*83.75)*3):100)*3
((6700*3):100)*3
(20100:100)*3

603 . . .

Now let's try with an easier map

one of incognito's perf

((((AR*10)*accuracy)*KM/2):100)*grade

((((5*10)*100)*16/2):100)*6
(((50*100)16/2):100)*6
((5000*8):100)*6
(40000:100)*6
400*6

2400pp where he actually gets 135 with actual pp system

just fo fun, let's see of the DT+HR apply to the AR in the equation :D

((((9*10)*100)*16/2):100)*6
(9000*8):100*6
72000:100*6
720*6

It gives a little 4320pp

That's all for me o/


The next time, you should MAYBE make the mods count for like...dunno...100 times less and add something concerning the star rating of the map...?

also make the "KM/2" => "(2xKM)/2" so nomod will really be x1 instead of x0.5 to the result, and with the mod value to at LEAST 50 times less, it will most likely make more sense.
vincaslt
101 posts
Joined January 2014
vincaslt
Are you saying that a player that has hundreds of full combos on hard maps but can barely pass 4.5 star insane is equally skilled with a player who has a dozen 6 star map full combos? I dont understand why you want to get pp for everything. Isn't score already like that? Current system forces you to keep improving to get pp, your system just rewards playing a lot of maps...
abraker
Global Moderator
8,327 posts
Joined July 2014
abraker

BlackWiddow wrote:

Tell me what you think about this! :D
The formula is STILL a shipwreck and apparently so is your mathematical knowledge. First, all you are doing is this:
0.05 * AR * Acc * KM * Grade

Just multiplying stuff together is already a bad intensive when it comes to comparing performance points. Sure they may be ranked the right way, but by no means will they be distributed properly. Also where the hell is the 6 in grade coming from???

Why is this receiving stars and why is this not an invalid request?
Jordanrere
Jordanrere
Some minutes ago i was in another player page, then i saw his top ranks... i saw one with 108pp, i donwload it and then i made an FC with the same mods on it but i didn't got any pp... why? someone explain me that.
Endie-
784 posts
Joined March 2015
Endie-

Jordanrere wrote:

Some minutes ago i was in another player page, then i saw his top ranks... i saw one with 108pp, i donwload it and then i made an FC with the same mods on it but i didn't got any pp... why? someone explain me that.
Accuracy
- e - v - b-
184 posts
Joined August 2014
- e - v - b-
With this formula a 1 star dthdhr ss would be worth more than Rafis' Best FriendS dthdhr play
Jordanrere
Jordanrere

erikG wrote:

Jordanrere wrote:

Some minutes ago i was in another player page, then i saw his top ranks... i saw one with 108pp, i donwload it and then i made an FC with the same mods on it but i didn't got any pp... why? someone explain me that.
Accuracy
Yes he got 100% and i got 95% but that diference it's too big 106(100%) to 0(95%)
Endaris
5,282 posts
Joined June 2010
Endaris
You're only playing easy maps with a lot of mods which is why the majority of the pp comes from accuracy.
Especially if the map was short you just got barely any accuracy-pp for 95% compared to 100% and the aim-pp wasn't enough.
You can easily confirm this with maps like this one:
https://osu.ppy.sh/b/187830
With HR you get 159pp for SS but only 61pp for 93% accuracy. Same story in your case.
Jordanrere
Jordanrere
Thanks i was lost to understand it, now it make sense to me.
Insanix_old_1
14 posts
Joined April 2015
Insanix_old_1
The best pp system which it is used in o!m
Lex TheGayOtter
24 posts
Joined August 2014
Lex TheGayOtter
Your system completely disregards OD and the fact that high AR doesn't make a map harder for most players, it's a pretty bad system compared to the one we currently have
You need to focus on improving instead of doing the same difficulty maps to try and get more PP
Improve then farm
In regards to the PP system, the one we have is fine albeit can be improved and has been improved over time and I believe that it will continue to be improved, it will always have it's flaws and exploits but it's the best thing we have
Remyria
361 posts
Joined June 2012
Remyria

abraker wrote:

BlackWiddow wrote:

Tell me what you think about this! :D
The formula is STILL a shipwreck and apparently so is your mathematical knowledge. First, all you are doing is this:
0.05 * AR * Acc * KM * Grade

Just multiplying stuff together is already a bad intensive when it comes to comparing performance points. Sure they may be ranked the right way, but by no means will they be distributed properly. Also where the hell is the 6 in grade coming from???

Why is this receiving stars and why is this not an invalid request?

1. read the post I made on the page 2, like 2-3 higher than yours that I'm answering to

2. probably SS=6 S=5 A=4 B=3 C=2 D=1 Of what I understood
casmith789
45 posts
Joined June 2013
casmith789
You can always use score rankings instead of PP if you want something that increases however much you play. At the end of each map you play, scroll down and you will see your score ranking. This will improve for every new ranked map you play, or if you improve your score on an old map. The top score rankings for this are at https://osu.ppy.sh/p/playerranking.

Your specific request would in my view make the scoring system worse as it doesn't take skill into account. While the current system isn't perfect it is still roughly correlated with skill.
Remyria
361 posts
Joined June 2012
Remyria

casmith789 wrote:

You can always use score rankings instead of PP if you want something that increases however much you play. At the end of each map you play, scroll down and you will see your score ranking. This will improve for every new ranked map you play, or if you improve your score on an old map. The top score rankings for this are at https://osu.ppy.sh/p/playerranking.

Your specific request would in my view make the scoring system worse as it doesn't take skill into account. While the current system isn't perfect it is still roughly correlated with skill.
the score ranking under the map result screen doesn't update, sadly
Please sign in to reply.

New reply

0
1 / 36