I believe you'd have to be a little more specific and base it on the misses/100s/50s count as well.PinkHusky wrote:
How big of a PP boost does a FC give? Since this is like osu!tp would it be the misses that are hurting me greatly? I thought that was something peppy was trying to do away with?
On the song 1 Year 2 Months 20 Days (Mapped by Athena Tennos) - Difficulty "Neko" Using DT only I got as follows.
Score: 2,156,908 (x295) 90.96% DT
Current PP: 1808
New Score: 3,012,150 (x276) 95.74% DT
PP Change: 1809
New Score: 5,971,551 (x569) 93.99% DT
PP Change: 1830
New Score: 6,055,706 (x570) 95.31% DT
PP Change: 1835
An almost 5% accuracy increase gave me 1 PP.
A Score only increase and 2% accuracy decrease gave me 21 PP.
Then a small increase of 1.5% accuracy gave me 5 after that.
Intended? No misses the way to go?
forum
Performance Points feedback and suggestions (Standard)
posted
Total Posts
2,750
Topic Starter
Combo is very relevant for PP. You've doubled your combo!PinkHusky wrote:
How big of a PP boost does a FC give? Since this is like osu!tp would it be the misses that are hurting me greatly? I thought that was something peppy was trying to do away with?
On the song 1 Year 2 Months 20 Days (Mapped by Athena Tennos) - Difficulty "Neko" Using DT only I got as follows.
Score: 2,156,908 (x295) 90.96% DT
Current PP: 1808
New Score: 3,012,150 (x276) 95.74% DT
PP Change: 1809
New Score: 5,971,551 (x569) 93.99% DT
PP Change: 1830
New Score: 6,055,706 (x570) 95.31% DT
PP Change: 1835
An almost 5% accuracy increase gave me 1 PP.
A Score only increase and 2% accuracy decrease gave me 21 PP.
Then a small increase of 1.5% accuracy gave me 5 after that.
Intended? No misses the way to go?
Is it a direct relation like (combo/max combo)*(Rest of PP calculation)? I have an almost-FC score that I don't think I'll ever beat in accuracy (and it's second in my top plays), so I'm not sure if it's even worth the effort, I don't wanna end up losing PP for the loss in acc + combo gain.
I think it depends on the length of the map and the actual possible combo. I had a DT play with 3 misses and 96% acc and when I improved to FC but ruined my acc(92%) I lost like 4 pp. It was 40 seconds on DT.nooblet wrote:
Is it a direct relation like (combo/max combo)*(Rest of PP calculation)? I have an almost-FC score that I don't think I'll ever beat in accuracy (and it's second in my top plays), so I'm not sure if it's even worth the effort, I don't wanna end up losing PP for the loss in acc + combo gain.
Is it just me, or are more slider pattern oriented maps seemed to have low star ratings in comparison to more circle patten oriented maps.
For example:
https://osu.ppy.sh/s/28425 (Black Rebel 3.43 stars) vs https://osu.ppy.sh/b/178966&m=0 (Talent Shredder 3.29 stars), but I am pretty sure most will agree that Talent Shredder is harder than Black Rebel.
Or take Skystar maps that frequently depend on you to leave sliders early to do jumps have lower star ratings than maps that depend on circle jumps
https://osu.ppy.sh/s/122658 (Maware (Skystar) 3.42 stars) and https://osu.ppy.sh/b/318294&m=0 (S.M.L (Skystar) 3.38 stars ) versus Fycho's https://osu.ppy.sh/s/122233 (Univer Page 3.33 stars) or https://osu.ppy.sh/b/318294&m=0 (S.M.L 3.52 stars )
(Sorry for the limited selection of songs, mostly just using the songs I was farming recently)
For example:
https://osu.ppy.sh/s/28425 (Black Rebel 3.43 stars) vs https://osu.ppy.sh/b/178966&m=0 (Talent Shredder 3.29 stars), but I am pretty sure most will agree that Talent Shredder is harder than Black Rebel.
Or take Skystar maps that frequently depend on you to leave sliders early to do jumps have lower star ratings than maps that depend on circle jumps
https://osu.ppy.sh/s/122658 (Maware (Skystar) 3.42 stars) and https://osu.ppy.sh/b/318294&m=0 (S.M.L (Skystar) 3.38 stars ) versus Fycho's https://osu.ppy.sh/s/122233 (Univer Page 3.33 stars) or https://osu.ppy.sh/b/318294&m=0 (S.M.L 3.52 stars )
(Sorry for the limited selection of songs, mostly just using the songs I was farming recently)
Topic Starter
The reason is, that the current way sliders are dealt with is very generous in terms of the slider's favor. Currently there is no measure as how hard it is to actually follow a slider. The minimum distance you have to move to complete the slider is added up to the jump to the next hitobject, but that's it atm. This "minimum distance" is required to not give fast repeatsliders, or even worse: slider-streams, ridiculous pp amounts.blissfulyoshi wrote:
Is it just me, or are more slider pattern oriented maps seemed to have low star ratings in comparison to more circle patten oriented maps.
For example:
https://osu.ppy.sh/s/28425 (Black Rebel 3.43 stars) vs https://osu.ppy.sh/b/178966&m=0 (Talent Shredder 3.29 stars), but I am pretty sure most will agree that Talent Shredder is harder than Black Rebel.
Or take Skystar maps that frequently depend on you to leave sliders early to do jumps have lower star ratings than maps that depend on circle jumps
https://osu.ppy.sh/s/122658 (Maware (Skystar) 3.42 stars) and https://osu.ppy.sh/b/318294&m=0 (S.M.L (Skystar) 3.38 stars ) versus Fycho's https://osu.ppy.sh/s/122233 (Univer Page 3.33 stars) or https://osu.ppy.sh/b/318294&m=0 (S.M.L 3.52 stars )
(Sorry for the limited selection of songs, mostly just using the songs I was farming recently)
I'll look into how this can be addressed without being too arbitrary.
Another reason at least for some of these maps to be underrated is, that they feature quick single passages with low spacing which gets underrated in the current algorithm. Couldn't really find a way to fix that without completely breaking spaced streams yet.
Holy shit, Tom (or whoever actually did this). Thank you for the pp values for scores on profile. I just spent a few minutes getting to the bottom, and felt so nostalgic.
Wow yes, this is greatLach wrote:
Holy shit, Tom (or whoever actually did this). Thank you for the pp values for scores on profile. I just spent a few minutes getting to the bottom, and felt so nostalgic.
Does the star difficulty in itself take into consideration the length of the song, or is it the length only taken into consideration when the PP gained ?
Example: 2 maps, exact same level of difficulty, but one is 3 minutes and the other is 6 minutes. Will they both be 3 stars, but give different amounts of PP, or will one be 3 stars and the other 3.2 stars ?
Example: 2 maps, exact same level of difficulty, but one is 3 minutes and the other is 6 minutes. Will they both be 3 stars, but give different amounts of PP, or will one be 3 stars and the other 3.2 stars ?
Displaying PP on webpage is awesome. I am loving the direction the ranking system is headed ever since the introduction of PPv2. Well done <3
I still suggest replacing the in-game star system with the web one (which I assume matches PPv2 'level' system). I have resorted to doing it manually
It's a lot of hassle to do, the in-game collection system is quite poor (messed up when a map belongs in more than one collection, requires external tool to sort). Integrating the PPv2 star system would also allow updating the level display based on selected mods. Additionally it would help people find maps to play that match their skill level. Currently there's only 2 star difficulties: "5 stars" and "don't bother", I imagine it's the same for many people. PPv2 star system makes a lot more sense where it's an actual spectrum.
I remember when I started playing and reached the point where hard maps were too easy, but I couldn't find suitable maps as most 5 star ones were too hard. Now I can FC many 5 star maps, but I still have a hard time finding good maps to play as most 5 star maps are still either too easy or too hard.
I still suggest replacing the in-game star system with the web one (which I assume matches PPv2 'level' system). I have resorted to doing it manually
It's a lot of hassle to do, the in-game collection system is quite poor (messed up when a map belongs in more than one collection, requires external tool to sort). Integrating the PPv2 star system would also allow updating the level display based on selected mods. Additionally it would help people find maps to play that match their skill level. Currently there's only 2 star difficulties: "5 stars" and "don't bother", I imagine it's the same for many people. PPv2 star system makes a lot more sense where it's an actual spectrum.
I remember when I started playing and reached the point where hard maps were too easy, but I couldn't find suitable maps as most 5 star ones were too hard. Now I can FC many 5 star maps, but I still have a hard time finding good maps to play as most 5 star maps are still either too easy or too hard.
I believe Tom has said he's working on it.
I just checked out those pp values and played a map that supposedly granted me 39 pp, the value increased to 52 yet my pp is still the same. Is this intended? If so, how does it work?
the pp shown on your profile is the raw pp amount, before the pp equation is applied.Kasugunai wrote:
I just checked out those pp values and played a map that supposedly granted me 39 pp, the value increased to 52 yet my pp is still the same. Is this intended? If so, how does it work?
also i have a suggestion; maybe show the raw pp a play is worth ingame on the score screen? to give an idea of how much the song would be worth.
It's still weird that a D-rank play (around 66% acc) is worth 52 raw pp when it's not even worth 1 after the equation is applied.
That 52 raw pp should be worth about 6pp (according to it's position on the rank list), but by making that score, any performance worth less than 52pp gives 5% less (except the scores that were worse than 39 raw pp, but those were giving practically zero anyways), so you might end up getting practically nothing.Kasugunai wrote:
It's still weird that a D-rank play (around 66% acc) is worth 52 raw pp when it's not even worth 1 after the equation is applied.
Topic Starter
Soon the percentage by how much a given score is weighted will be shown along with the raw pp value to avoid confusion. 
I approve of this.Soulg wrote:
also i have a suggestion; maybe show the raw pp a play is worth ingame on the score screen? to give an idea of how much the song would be worth.
New update, showing PP super, but can you add information about mod which played map (HR, DT, DT+HD, None...), and "new" as in the TP system. It is simply amazing
Any plans to implement a raw PP values into the "Recent Plays" area of a profile? Or something equal to that?Tom94 wrote:
Soon the percentage by how much a given score is weighted will be shown along with the raw pp value to avoid confusion.
And Accuracy as well would be nice. When I go over my top ranks I find myself looking for low accuracy plays to improve upon.Fanker wrote:
New update, showing PP super, but can you add information about mod which played map (HR, DT, DT+HD, None...), and "new" as in the TP system. It is simply amazing
PP scores on profile are awesome.
Sorry if this sounds dumb, but why is it that only one map is weighted at 100% of the base pp value for everyone?
I would have thought that there would be maps that are slightly easier than the top one which would be weighted 100% but not give as much pp because, well, it's easier, while the top map would be weighted more around 95-100% based on your results.
Either that or my idea of how the pp is being weighted is skewed, and if so then please explain ;-;
I would have thought that there would be maps that are slightly easier than the top one which would be weighted 100% but not give as much pp because, well, it's easier, while the top map would be weighted more around 95-100% based on your results.
Either that or my idea of how the pp is being weighted is skewed, and if so then please explain ;-;
Topic Starter
The weight is only based on the position of the score's pp value compared to the other pp values of the player's scores. This basically ensures that there is no "farming" or whatever you may call it. In other words: Only your best pp scores count towards your final pp.TMoI wrote:
Sorry if this sounds dumb, but why is it that only one map is weighted at 100% of the base pp value for everyone?
I would have thought that there would be maps that are slightly easier than the top one which would be weighted 100% but not give as much pp because, well, it's easier, while the top map would be weighted more around 95-100% based on your results.
Either that or my idea of how the pp is being weighted is skewed, and if so then please explain ;-;
It has nothing to do with map difficulty. If it had, then someone could keep playing maps of the same difficulty and keep getting pp for that, which would be quite contrary to what pp wants to encourage: Play harder maps, improve.
So for us to get pp we need to get a score that is higher then the 104 top scores but doing that it will make the top players win less and less because its not every day maps really hard are madeTom94 wrote:
The weight is only based on the position of the score's pp value compared to the other pp values of the player's scores. This basically ensures that there is no "farming" or whatever you may call it. In other words: Only your best pp scores count towards your final pp.TMoI wrote:
Sorry if this sounds dumb, but why is it that only one map is weighted at 100% of the base pp value for everyone?
I would have thought that there would be maps that are slightly easier than the top one which would be weighted 100% but not give as much pp because, well, it's easier, while the top map would be weighted more around 95-100% based on your results.
Either that or my idea of how the pp is being weighted is skewed, and if so then please explain ;-;
It has nothing to do with map difficulty. If it had, then someone could keep playing maps of the same difficulty and keep getting pp for that, which would be quite contrary to what pp wants to encourage: Play harder maps, improve.
More mods!snosey wrote:
So for us to get pp we need to get a score that is higher then the 104 top scores but doing that it will make the top players win less and less because its not every day maps really hard are made
rrtyui is that far up because he DTHD'd several maps that only have HRHD in the top rankings.
And at that point, I'm assuming that it's kinda hard to rank because everyone there is just as good as you, you know?
i never said i was good you are the one who is assuming lol i just said it would be hard and like you said rrtyui did amazing ranks but he will need to continue doing dthd more maps that most people do hrhd but of course he also has maps that are hell hard to fc without modsTMoI wrote:
More mods!snosey wrote:
So for us to get pp we need to get a score that is higher then the 104 top scores but doing that it will make the top players win less and less because its not every day maps really hard are made
rrtyui is that far up because he DTHD'd several maps that only have HRHD in the top rankings.
And at that point, I'm assuming that it's kinda hard to rank because everyone there is just as good as you, you know?
He was answering your question, not assuming anything about you. You were the one assuming he was assuming.snosey wrote:
i never said i was good you are the one who is assuming lol
It is hard. If it was easy, it wouldn't be impressive to be the #1. The goal is to make it hard, so that the top player deserves their position. I don't see what's bad about this.snosey wrote:
i just said it would be hard and like you said rrtyui did amazing ranks but he will need to continue doing dthd more maps that most people do hrhd but of course he also has maps that are hell hard to fc without mods
They are supposed to win less for playing maps that are not up to their standard. Did you even bother reading what you just wrote? They get such small amounts of pp for those maps because they are capable of doing much harder ones. That's how the system is supposed to work. They're not playing to get pp for playing maps they can do without even breaking a sweat. They, like everyone else, are supposed to improve to go further in pp. They have already gained all the pp that "other people are getting for doing hdhr", so don't compare it to that.snosey wrote:
So for us to get pp we need to get a score that is higher then the 104 top scores but doing that it will make the top players win less and less because its not every day maps really hard are made
i interpreted he/her answer wrong thank you for clarifying me-Scylla- wrote:
He was answering your question, not assuming anything about you. You were the one assuming he was assuming.snosey wrote:
i never said i was good you are the one who is assuming lolIt is hard. If it was easy, it wouldn't be impressive to be the #1. The goal is to make it hard, so that the top player deserves their position. I don't see what's bad about this.snosey wrote:
i just said it would be hard and like you said rrtyui did amazing ranks but he will need to continue doing dthd more maps that most people do hrhd but of course he also has maps that are hell hard to fc without mods
"Draxuss wrote:
They are supposed to win less for playing maps that are not up to their standard. Did you even bother reading what you just wrote? They get such small amounts of pp for those maps because they are capable of doing much harder ones. That's how the system is supposed to work. They're not playing to get pp for playing maps they can do without even breaking a sweat. They, like everyone else, are supposed to improve to go further in pp.snosey wrote:
So for us to get pp we need to get a score that is higher then the 104 top scores but doing that it will make the top players win less and less because its not every day maps really hard are made
what you just said was what i wrote just without the "without even breaking a sweat" they will have to do harder maps but after some time the maps that don't give much pp are still really hard this is what i wanted to say but forget it let time pass by and see were this system will go have fun playing
Wait so, how do the percentages work?
Does it work like if you have scores worth...
1st rank 100pp - 100% = 100pp
2nd rank100pp - 95% = 95pp
Totaling up to a total of 195pp with just 2 scores?
Does it work like if you have scores worth...
1st rank 100pp - 100% = 100pp
2nd rank100pp - 95% = 95pp
Totaling up to a total of 195pp with just 2 scores?
That's how I'm assuming it works, and that's what makes the most sense. I counted all my scores with >0% weighting, and came down to 104 weighted scores. I wonder if this is the same for everyone, or if that varies as well.pop102 wrote:
Wait so, how do the percentages work?
Does it work like if you have scores worth...
1st rank 100pp - 100% = 100pp
2nd rank100pp - 95% = 95pp
Totaling up to a total of 195pp with just 2 scores?
Yes.pop102 wrote:
Wait so, how do the percentages work?
Does it work like if you have scores worth...
1st rank 100pp - 100% = 100pp
2nd rank100pp - 95% = 95pp
Totaling up to a total of 195pp with just 2 scores?
In that case, if all your scores are worth 100 raw pp (and assuming you got many of them), then your total pp would be about 2000pp, and you won't ever be able to get more than 2000pp total unless you get scores that are worth over 100 raw pp.
The 104th score should have a weighting of 0.482231% (it shows as 0% because the value is rounded).-Scylla- wrote:
That's how I'm assuming it works, and that's what makes the most sense. I counted all my scores with >0% weighting, and came down to 104 weighted scores. I wonder if this is the same for everyone, or if that varies as well.
There's no bottom limit to the formula. Theoretically you could get an unlimited amount of 100pp scores and each score would add something to your pp. Only after awhile the pp given is so small that it could take 100s of performances to get a single pp from it.Full Tablet wrote:
Yes.pop102 wrote:
Wait so, how do the percentages work?
Does it work like if you have scores worth...
1st rank 100pp - 100% = 100pp
2nd rank100pp - 95% = 95pp
Totaling up to a total of 195pp with just 2 scores?
In that case, if all your scores are worth 100 raw pp (and assuming you got many of them), then your total pp would be about 2000pp, and you won't ever be able to get more than 2000pp total unless you get scores that are worth over 100 raw pp.
The function does in fact converge, so you'll never be able to surpass 2000pp unless you get at least one score with a pp value of 101 or better (Assuming the current decay value of 0.95)
The thing is, not even with an infinite amount of 100 raw pp scores your total pp will past over 2000.TheVileOne wrote:
There's no bottom limit to the formula. Theoretically you could get an unlimited amount of 100pp scores and each score would add something to your pp. Only after awhile the pp given is so small that it could take 100s of performances to get a single pp from it.
Theoretically:
10 100pp scores: 862.4 total pp.
100 100pp scores: 1988.75 total pp.
1,000 100pp scores: 2000 - 10^(-19) total pp.
100,000 100pp scores: 2000 - 4.36*10^(-2225) total pp.
The values in practice could be different because of rounding errors with floating-point calculations, but even then the errors would be incredibly small.
Then I undercounted, since I started counting at 1%. Would it be safe to assume that 110 scores are counted, or would this still vary per player?Full Tablet wrote:
The 104th score should have a weighting of 0.482231% (it shows as 0% because the value is rounded).-Scylla- wrote:
That's how I'm assuming it works, and that's what makes the most sense. I counted all my scores with >0% weighting, and came down to 104 weighted scores. I wonder if this is the same for everyone, or if that varies as well.
In theory, all scores are counted.-Scylla- wrote:
Then I undercounted, since I started counting at 1%. Would it be safe to assume that 110 scores are counted, or would this still vary per player?
For performance reasons, there could be a cut-off rank (since after so many ranks, each score adds a meaningless amount to the total).
In case you guys are interested in the maths behind all of this (for example why it's impossible to break certain pp barriers), it's called a "geometric series". You can find lots of online resources with information on the topic.
Well I guess that it could approach any value without ever reaching it. What makes you say that value is 2000 and not some other number?Full Tablet wrote:
The thing is, not even with an infinite amount of 100 raw pp scores your total pp will past over 2000.TheVileOne wrote:
There's no bottom limit to the formula. Theoretically you could get an unlimited amount of 100pp scores and each score would add something to your pp. Only after awhile the pp given is so small that it could take 100s of performances to get a single pp from it.
Theoretically:
10 100pp scores: 862.4 total pp.
100 100pp scores: 1988.75 total pp.
1,000 100pp scores: 2000 - 10^(-19) total pp.
100,000 100pp scores: 2000 - 4.36*10^(-2225) total pp.
The values in practice could be different because of rounding errors with floating-point calculations, but even then the errors would be incredibly small.