forum

How to handle PP with Mods! (Idea)

posted
Total Posts
11
This is a feature request. Feature requests can be voted up by supporters.
Current Priority: +28
Topic Starter
NekomusuMae
I no longer agree with this post.
March 2019 Version of this Post

Multiple mods have weird and over complicated calculations. My idea is an easier and MORE SENSICAL way of going about Scoring, Star Rating, and pp spreading maps with these difficulties.

General Concept


In osu!, mods change one or more variables that can be set in the mapmaker.
What I say is simple and easy to understand. If I were to play a map half-time, I would receive a 0.3x multiplier on my score. Okay. But if I got (let's say) 10,000 score on a map with half time and then compared that to a modified version of the same map where I manually changed the map files and settings to where it was exactly the same as playing said map on HT and played it the exact same (making the exacts same cursor and keyboard movements) I would get a COMPLETELY different score! What I say is this. If a player plays a map with a mod, score, pp, ect. should be exactly the same as playing it with said modifications applied manually.

Example per Mod


  1. EZ (Easy) - Maps should be treated as if every setting was halved with the exception of HP Drain which should be halved, then divided by 3 to reflect the 3 health bars received in EZ.
  2. NF (No Fail) - Maps should be treated EXACTLY THE SAME if the health bar never reaches 0. If the health bar does reach 0, take every bit of HP the would have gone negative (if that was possible) and treat map as how much Health Drain would be required to make a play like that possible without EZ.
  3. HT (Half Time) - Treat map as if it was 75% it's actual speed, and all other settings were slightly lower. I would be more specific but the wiki is very unclear as to what else HT does.
  4. HR (Hard Rock) - Maps should be treated as if CS was raised by 30%, all other settings were raised by 40%, and the map was flipped over the X-axis.
  5. SD (Sudden Death) - NO CHANGE
  6. PF (Perfect) - NO CHANGE
  7. DT (Double Time) - Maps should be treated as if maded 150% faster, and all other settings were slightly slower. I would be more specific but the wiki is very unclear as to what else DT does.
  8. NC (Nightcore) - SAME AS DT
  9. HD (Hidden) - Can't really say about this because this is a lot more complicated than everything else. Could be treated as if the Approach Rate was much shorter or something.
  10. FL (Flashlight) - SAME AS HD
  11. SO (Spun Out) - Could be treated as if spinner was missed when it comes to score and such, but no punishment to HP and % and stuff.

TL:DR


Basically, treat maps that are faster as if they were mapped in that way. A map with Nightcore enabled should be no different from a map of a Nightcore song.
abraker

MLisDreaming wrote:

If I were to play a map half-time, I would receive a 0.3x multiplier on my score. Okay. But if I got (let's say) 10,000 score on a map with half time and then compared that to a modified version of the same map where I manually changed the map files and settings to where it was exactly the same as playing said map on HT and played it the exact same (making the exacts same cursor and keyboard movements) I would get a COMPLETELY different score! What I say is this. If a player plays a map with a mod, score, pp, ect. should be exactly the same as playing it with said modifications applied manually.
sigh

I agree with you, but you have to understand why it the way it is right now. Currently various mods are mixed together in the same leaderboard. If there were no mod multipliers, then HT plays would tend to get higher scores than no-mod and DT because it is easier (ignoring AR shenanigans). DT plays would tend to get lower scores because it is harder than no-mod and HT. Some goes for various other mods. How unfair would it be to play DT and not being able to beat your no-mod score because it's much much harder? The score multipliers are a wonky thing, but they are meant to reflect some sort of justification to allow harder scores reach the leaderboards higher than stuff that is easier and vice versa.

For this problem to truly disappear, scores from one mod should be compared to scores from the same mod. You can already do that when you filter to mod specific scores . . . if you are supporter. I think you should start to see the issue with how the system is set up. By default we look at a leaderboard that mixes various mods while mod specific leaderboards are exclusive to supporters, however to be able to allow what you propose you should be looking a mod specific leaderboards by default. So that's the dilemma. You need to make mod specific leaderboards available to everyone and make that the default to get rid of mod multipliers.
Topic Starter
NekomusuMae

abraker wrote:

...however to be able to allow what you propose you should be looking a mod specific leaderboards by default. So that's the dilemma. You need to make mod specific leaderboards available to everyone and make that the default to get rid of mod multipliers.

I feel that isn't a solution. Just avoiding the problem. If my idea was put in and people were getting better scores with half-time. I don't feel like that's a problem with my suggestion. I feel like that's a problem with the way maps are score-rated.
And your right. You get more score from doing good on a easy map than doing decent on a hard map. But you shouldn't. On the Score-based Board (https://osu.ppy.sh/rankings/osu/score) most top people just play easy maps. That's why we moved to pp from Score way back as our leaderboard-system. But we never changed how our map leaderboards worked. The problem is our score system for individual maps. Either change it from Score to pp, or fix the score system.

And btw, before someone mentions it, I realize Score V2 would completely break my idea.
abraker
At that point we can get rid of score entirely, and make pp the new score
Topic Starter
NekomusuMae

abraker wrote:

If there were no mod multipliers, then HT plays would tend to get higher scores than no-mod and DT because it is easier (ignoring AR shenanigans). DT plays would tend to get lower scores because it is harder than no-mod and HT. Some goes for various other mods. How unfair would it be to play DT and not being able to beat your no-mod score because it's much much harder? The score multipliers are a wonky thing, but they are meant to reflect some sort of justification to allow harder scores reach the leaderboards higher than stuff that is easier and vice versa.

I feel like this is a problem not with DT and HT being easier and harder. Just that the score system has lots of problems.

abraker wrote:

At that point we can get rid of score entirely, and make pp the new score

And the pp system has problems. I feel like osu! needs a complete reworking with how everything is scored and how leaderboards work.
xenal

MLisDreaming wrote:

You get more score from doing good on a easy map than doing decent on a hard map. But you shouldn't. On the Score-based Board (https://osu.ppy.sh/rankings/osu/score) most top people just play easy maps.

Score is based from FC, aka large combos (then accuracy and mods comes in slightly) and it just happens that most easy maps are easier than hard maps, mhm I wonder why. The score leaderboard is actually based from playing all maps in the game (as shown by xaxuma in the top by this large of a margin as he did all of them). And if you think "easy being easier than hard" was the reason why pp was introduced, you are wrong (it's because score doesn't reflect skill, just time wasted clicking circles... sigh...)

Leaderboard are not just there for score, but to show the most impressive scores on a map. HT scores being the same as DT scores while DT is more impressive breaks this concept.

Abraker first anwser should have settle this.
abraker

xenal wrote:

Score is based from FC, aka large combos (then accuracy and mods comes in slightly)
While true for the standard and ctb gamemodes, remember there is mania and taiko as well.

xenal wrote:

Leaderboard are not just there for score, but to show the most impressive scores on a map.
Leaderboards mean different things for different people. For you it's a show of how impressive the score is. For me it's a relative measure (if I get worse than quarter or eighth my rank on a mania map, then I know I can do better).
Merami Fox

abraker wrote:

MLisDreaming wrote:

If I were to play a map half-time, I would receive a 0.3x multiplier on my score. Okay. But if I got (let's say) 10,000 score on a map with half time and then compared that to a modified version of the same map where I manually changed the map files and settings to where it was exactly the same as playing said map on HT and played it the exact same (making the exacts same cursor and keyboard movements) I would get a COMPLETELY different score! What I say is this. If a player plays a map with a mod, score, pp, ect. should be exactly the same as playing it with said modifications applied manually.
sigh

I agree with you, but you have to understand why it the way it is right now. Currently various mods are mixed together in the same leaderboard. If there were no mod multipliers, then HT plays would tend to get higher scores than no-mod and DT because it is easier (ignoring AR shenanigans). DT plays would tend to get lower scores because it is harder than no-mod and HT. Some goes for various other mods. How unfair would it be to play DT and not being able to beat your no-mod score because it's much much harder? The score multipliers are a wonky thing, but they are meant to reflect some sort of justification to allow harder scores reach the leaderboards higher than stuff that is easier and vice versa.

For this problem to truly disappear, scores from one mod should be compared to scores from the same mod. You can already do that when you filter to mod specific scores . . . if you are supporter. I think you should start to see the issue with how the system is set up. By default we look at a leaderboard that mixes various mods while mod specific leaderboards are exclusive to supporters, however to be able to allow what you propose you should be looking a mod specific leaderboards by default. So that's the dilemma. You need to make mod specific leaderboards available to everyone and make that the default to get rid of mod multipliers.






um, so the way im getting the original post is like this, dt and ht score should reflect the actual speed of it, your really just proving him right by saying such thing, assuming i am understanding this correctly of coarse
abraker

bommy985 wrote:

um, so the way im getting the original post is like this, dt and ht score should reflect the actual speed of it, your really just proving him right by saying such thing, assuming i am understanding this correctly of coarse
It's more complicated than that. He says,

MLisDreaming wrote:

You get more score from doing good on a easy map than doing decent on a hard map. But you shouldn't. On the Score-based Board (https://osu.ppy.sh/rankings/osu/score) most top people just play easy maps. That's why we moved to pp from Score way back as our leaderboard-system. But we never changed how our map leaderboards worked. The problem is our score system for individual maps. Either change it from Score to pp, or fix the score system.
Basically he argues that score is not being scaled correctly in respect to skill required. It's a very, very complicated issue to fix and we are better off just not mixing scores from different mods altogether unless someone can offer a way to resolve that and prove it works before application.
chokonmainatsu
few days ago i played FL for the first time and tried to fc a 4.1* 30 sec map. and it was hard af, i needed about 160 plays. while HD depends on every person. i can't play HD very well (3.5 stars with HR is the max for me) while other friends can easily play up to 5.99* with HD, some of them play worse without HD.

TLDR FL is a lot harder than HD and should be worth more
abraker
There are issues in trying to quantify difficulty due to memory in relation to skills such as speed, stamina, aim etc
Please sign in to reply.

New reply