forum

Performance Points feedback and suggestions (Standard)

posted
Total Posts
2,750
show more
uzzi

Symqn wrote:

Just got random +20pp and then random -10pp and -0,10 acc after finishing a hard map that i was playing for fun is there a penalty for playing easyer maps now?
Not sure, but I gained some pp and lost some accuracy as well from playing.
TheBigA
I would be interested in how sliderbreaks are treated (combo breaking ones and just -1 combo ones).
XGeneral2000

Symqn wrote:

Just got random +20pp and then random -10pp and -0,10 acc after finishing a hard map that i was playing for fun is there a penalty for playing easyer maps now?
I think there's a lot of tweaking/adding/updating going on right now, my pp increased this morning despite me not having even opened the client for over 24 hours. My "Top Performances" changed as well. I'm guessing we should give it some time to finish processing and stabilize.

With regards to the highest-performance vs highest-score problem, would it be possible to run a pp-value check after every play (regardless of score) and compare it to the pp-value of the current high score? And if the new play is a higher value, to update the pp accordingly even if high score doesn't change?

I admittedly don't know much about what calculations are done client-side vs server-side, or whether pp is calculated from scratch after every play, or anything like that (I'd be very interested to know, actually, if you have time and if that's disclosable), so I can't suggest much in terms of efficiency. My guess is that there's a table of high scores for every player on the server, and that pp is recalculated each time this table is updated...in this case, the only way to implement this higher-pp vs higher-score idea would be to store a new table of highest-pp scores for every player, which would be very memory-expensive (as I think you mentioned somewhere earlier). The only way I can see to get around this would be to migrate the pp calculations to the client-side, but I don't know how feasible that would be.
Natteke
I've tried to see if I can get any pp at all. Played a bunch of maps, mostly newly ranked with streams and jumps. Got decent acc and ranks, ended up with 0pp. Can someone explain what I'm doing wrong? Is it not real time? I'm confused.
Ippikiryu
Are all the older maps calculated into the current scores? Or are they still updating? Or do you have to replay some to get them reincluded?
Pettanko

Natteke wrote:

I've tried to see if I can get any pp at all. Played a bunch of maps, mostly newly ranked with streams and jumps. Got decent acc and ranks, ended up with 0pp. Can someone explain what I'm doing wrong? Is it not real time? I'm confused.
I think it takes time before your pp updates because when I got a top performance a few days ago it took some time before it showed in my performace history. I'm guessing it updates every few hours or so?
969363565

Draxuss wrote:

Most, if not all of your flashlight scores are apparently normal and easy maps, while the ones in top performance are insane and hard. System working as planned?
I don't know how it works and what should working as planned be like. I'm sure that I have scores that are insane +FL and hard +DTFL, but they are still defeated by the none scores.

Rewben2 wrote:

Tom94 wrote:

FL gives a huge bonus on maps which are hard to aim. I am tallking about a 36% increase. However if the movement of the map you played is not that complicated, then even FL won't be able to pull it up by much. I've already been considering to increase FL's bonus even further in the past, since it makes aiming so much harder.
Thanks a lot. I hadn't read the early replies. :lol:
However, if flashlight can only get a huge bonus on maps which are hard to aim, should DT get a huge bonus only in maps which have high BPM? I don't think DT is much harder than FL, especially in some slow and long maps.


By the way, how about the maps which don't have that many streams and jumps but are hard to read? Is this kind of difficulty considered?
XGeneral2000

969363565 wrote:

However, if flashlight can only get a huge bonus on maps which are hard to aim, should DT get a huge bonus only in maps which have high BPM? I don't think DT is much harder than FL, especially in some slow and long maps.
Theoretically, DT should already only give huge bonus in maps which are harder to begin with. It doesn't add a % multiplier - the difficulty is recalculated from scratch with DT in effect - the same as if you were playing a version mapped at 1.5x speed with no mods. So if a map is slow to begin with, the difficulty won't be that high even with DT.
Metro
Shouldn't country rankings be ordered by average performance?
Full Tablet

Metro wrote:

Shouldn't country rankings be ordered by average performance?

That would give too much ranking to some countries with only 1 player
NixXSkate

Tom94 wrote:

Kert wrote:

And also FL?
FL is not favoured at all?
http://osu.ppy.sh/s/18156 - this must be one of my top plays, yet it's nowhere close
(my assumption is results are ordered almost in the same way as in tp)
FL gives a huge bonus on maps which are hard to aim. I am tallking about a 36% increase. However if the movement of the map you played is not that complicated, then even FL won't be able to pull it up by much. I've already been considering to increase FL's bonus even further in the past, since it makes aiming so much harder.
I think the speed of the map (1/2 notes) with how jumpy it is should play a big part in the FL's bonus system.
http://osutp.net/scores?bid=154889
This is one of the most impressive FL scores yet due to the speed of the map along with its jumps (normally FL scores are on fairly slow maps), but we can see that the system didn't treat Dungeon kindly in comparison to the competition.

Sorry if it was mentioned before, the thread grew quite a bit and didn't want to look through all 17 pages. :o
Metro

Full Tablet wrote:

Metro wrote:

Shouldn't country rankings be ordered by average performance?

That would give too much ranking to some countries with only 1 player
True. But the country rankings are much different than in osu!tp, that's why I'm asking.
wyedwardchan

dedaloodak wrote:

4k# >> 12k#
gg peppy and crew -.-
I know that feel bro
Mine 5k#>>>>18k#

Just fun playing is now not an option, just keep retrying your insane maps dude :roll: :roll:
UNIFO
I have some concerns regarding Performance Points and tying it to difficulty of the map.

Most notably.... songs that involve 5.00 difficulty.

There are some songs at 5.00 difficulty that are not as difficult as some of the insane 5.00 difficulty maps, yet I believe they are treated the same for calculations for PP purposes. (I'll provide examples if needed).

This can be "set up for some level of exploit of getting PP" at around the 1k to 2k PP range, in my opinion. Ie... they can effectively raise their PP by playing and doing really well at the easy 5.00 difficulty maps.


Maybe we need to start ranking maps with 6.00 difficulty to really differentiate the hard and really hard maps... because there is a huge saturation of maps at the 4.75 - 5.00 difficulty range?
Wishy
Lol@people crying because they went from 4k to 600k. If you play for fun then you shouldn't care about your rank.

UNIFO, there is another difficulty calculation system, just ignore the stars, nobody cares about them, pp uses another system to evaluate maps, stars are there just because.
buny

Wishy wrote:

Lol@people crying because they went from 4k to 600k. If you play for fun then you shouldn't care about your rank.

UNIFO, there is another difficulty calculation system, just ignore the stars, nobody cares about them, pp uses another system to evaluate maps, stars are there just because.
i still have scores that took 1 pc in my best performance though
Wishy
Most of the scores I got on my best performance came from score that took me one play, your point? Some examples would be helpful.
buny
my point is that the weighings can still be improved
Wishy
Certainly.
UNIFO

Wishy wrote:

Lol@people crying because they went from 4k to 600k. If you play for fun then you shouldn't care about your rank.

UNIFO, there is another difficulty calculation system, just ignore the stars, nobody cares about them, pp uses another system to evaluate maps, stars are there just because.
Here is some definitive evidence/example that will leave anybody baffled on how the PP system would work.

Example 1:

Before improving the score of this map (https://osu.ppy.sh/b/183511), I had a score of like 700k on it, and it was considered one of my top performing songs.

Now compare that to the score of this map (https://osu.ppy.sh/b/312770) which still is at 1.5 Mil. I think this map is harder than the previous map, yet this was not one of my top performing song.

Note that both songs are at 5.00 difficulty. The 2nd one is the more difficult song. DESPITE this fact, my 700k score was a top performer (as of yesterday, now has improved). Now how does that make any sense what so ever?

Example 2:

https://osu.ppy.sh/b/169090?m=0 -> my top score is 1.4 Mil, 79% accuracy, max combo 170, and difficulty of 4.94. This song is in my top performer.

https://osu.ppy.sh/b/309501 -> my top score is 4.6 Mil, 87% accuracy, max combo 460, and difficulty of 5.00. This song NOT my top performer.

Again the 2nd one I feel is much more difficult. Yet somehow the first one is part of my top 10 performance, and the 2nd isn't.

----

I'm aware there are other factors that involve PP, such as how competitive the song could be. However, these two examples does show that something is kinda fishy occurring with 5.00 difficulty songs, and can be set up for exploit with specific songs.
Wishy
I'm too lazy to check out the maps, but just by looking at the scoreboard I can say what you usually consider to be easier is actually harder.

Didn't check any of the maps tho so I may be wrong. Star system has nothing to do anyways, trust me, just ignore it. You won't get a lot of pp from a random 5* map compared to hard 5* maps.
HoriKyoko
This feels a little stupid but, is the top 10 best performance in order of amount of pp awarded?
Lach

lucari386 wrote:

This feels a little stupid but, is the top 10 best performance in order of amount of pp awarded?
Yes. The higher ones are more valuable.
thelewa

dedaloodak wrote:

4k# >> 12k#
gg peppy and crew -.-
#2 -> #4 as you can see 4 is double the amount of 2 meaning that at a percentual level my rank dropped almost as much as yours did

GG NO RE THIS SYSTEM IS DOOMED




on a more serious note people who are complaining about their rank dropping should consider the possibility that there might be players better than them getting better scores than what they're getting

just saying that it is an almost certain possibility
Kinji
what i dun get is, how does player which has worse scores than me, performance is also not as good( not to brag), but have higher pp than me...
Rewben2

Kinji wrote:

what i dun get is, how does player which has worse scores than me, performance is also not as good( not to brag), but have higher pp than me...
Example? The system calculates their scores are worth more.
darkmiz

Kinji wrote:

what i dun get is, how does player which has worse scores than me, performance is also not as good( not to brag), but have higher pp than me...
check their best performance scores, and beat them because you think you're better. free pp
thelewa

Kinji wrote:

what i dun get is, how does player which has worse scores than me, performance is also not as good( not to brag), but have higher pp than me...
instead of asking about it you should provide names and examples of the scores

I also think that I should be #1 but because I'm not this system is bad
Kinji
click on my userpage, and this userpage. http://osu.ppy.sh/u/2621067, i have been 'stalking' his userpage to compare scores, and get btr, but my pp is always lower than him....
plaatinum

Kinji wrote:

click on my userpage, and this userpage. http://osu.ppy.sh/u/2621067, i have been 'stalking' his userpage to compare scores, and get btr, but my pp is always lower than him....
You have to get a better score than him on these maps.
http://osu.ppy.sh/b/226839?m=0
http://osu.ppy.sh/b/169450?m=0
http://osu.ppy.sh/b/252000?m=0
http://osu.ppy.sh/b/136400?m=0
http://osu.ppy.sh/b/169090?m=0
Kinji

Plaatinum wrote:

You have to get a better score than him on these maps.
http://osu.ppy.sh/b/226839?m=0
http://osu.ppy.sh/b/169450?m=0
http://osu.ppy.sh/b/252000?m=0
http://osu.ppy.sh/b/136400?m=0
http://osu.ppy.sh/b/169090?m=0
so basically i get btr score in this few maps, my pp will increase to the same as him? and how do i know that I have a btr score in that song than him? Thank you very much for answering my question. :D
AmaiHachimitsu
Just don't think "better score" means "being higher in scoreboard" only in this case. You need to use the same mods (or add some) and get better accuracy.

i.e. Your score won't be "better" if you beat DT score with FL
buny
also less misses
Lach
This guy also has 5k more playcount than you, Kinji, and probably has numerous other scores that obviously didn't make it into the list, but are of similar value.
Kinji
I see....thank you all for your comments, hardcore osu now xD
Wishy
I've seen how the system actually appreciates good scores with a random dumb miss (a few of my top performances are like that), therefore I can assume the system doesn't punish not-FCing that much. Now, the question, what would happen if you manage to clear something impossible, let's say you clear Airman on DT with a B, how much would that be worth? I mean that certainly requires an insane amount of skill and is probably harder than getting some random few misses score on the same map without any mod.

I know Airman is not a very good example because of low drain, but the point is quite clear.
buny

Tom94 wrote:

Misses are penalized very strongly if you have more than just a few. Your combo relative to the maximum possible combo also plays a huge role. But you can indeed get a huge amount of pp for lower scores. A prime example would be jesus1412's C score on https://osu.ppy.sh/b/124501?m=0 . It is a 86% 500 combo with DoubleTime in case you are curious. That's an insane score on such a map, you are free to look at Auto-mode play it with DoubleTime.

SS does not have a huge bonus by itself, but due to the high accuracy required to get one (100% is quite high, eh? :p) it indirectly is worth a lot.

Accuracy is still very relevant, however not as much as in ppv1 anymore.
Wishy
Thanks. So the more misses you get the harsher it gets (1 miss takes x pp, 2 misses takes 3x pp, 3 misses take 8x pp, etc) and the closer your maximum combo is to the maximum map's combo the better.

Damn I was hoping to fill my top performances with Cs.
buny
technically you can, but the amount of speed and aim that is needed to carry the overall pp anchored by the accuracy factor is probably either on par or higher than rrtyui

since pp uses only the end results, perhaps tom could make a program that can calculate pp gained from arbitrary input values
Symqn
19k>40k yea i guess kinda pp farm when i think about it after i raged about the drop this system is good only needs to fix bugs and real time update's.I have a suggestion it would be cool to be somewhere written how much pp and acc you gain or lose for a song,lets say under historical tab there would be
<song name> +9pp +0,03 acc or +2pp -0,01 acc.
show more
Please sign in to reply.

New reply