forum

The new score system

posted
Total Posts
22
This is a feature request. Feature requests can be voted up by supporters.
Current Priority: +4
Topic Starter
Abestado
I don't know if here is the right place to post it. But I'm noob in the forum, so I'm gonna let it here =S

I searched a little about this score v2 and the idea of changing the score system is good, but the new score itself isn't quite good.
(I'm not an english native speaker. I apologize for my grammar mistakes.)

I have an idea for score v2, but it kind of has nothing to do with the score v2 that peppy wants for the game.

So... My idea is to have a score based just on accuracy. It would be more "real" for the players. Giving the fact that keep a great combo doesn't mean you are a good player. It just means you are someone who have a lot of time to play and build consistency. The only way to build consistency is playing, and not everyone has so much time to play... I'll give an example:

Rohulk is a good accuracy player, and he can hold great combos. But to do that, I bet he had days where he played almost 10 hours straight. I can't play that much to build a good consistency. But I can play enough to have a good accuracy in some songs. And it doesn't take much time.

I'm not trying to change the system to be more comfortable for me, but for everyone... Everybody has that hard times where they can't FC nothing AT ALL. And with an ACC based score, that "hard time" would never come. Because you'll always be able to get a good acc in some songs.

It would be almost like mania, but the difference is that combo wouldn't count (in mania it does count a little). In order to make a good PP play, you would need just accuracy.

What's my point? - If you have a song that has 500 objects, and the player who played this song couldn't fc, but he managed to click on 499 objects, and missed only the last one.
If he had missed the first object, or one circle in the middle of the song, this wouldn't change his final pp amount for that song. It wouldn't matter which part of the song he missed. (I don't think is clear enough, but I think you can understand what I mean.)


Combo would be like a "fame" thing. So who has the higher combo, would be higher in the ranking screen of the song. But who couldn't get a huge combo, would have almost the same pp got from that player who got an fc (If he had the same quantity of accuracy).

I don't think this "system" that i'm talking about has gaps because if you miss a note, you automatically will lose acc, thus giving you less pp for your miss. It's like the system that we have today. It would change only the combo thing.

Making a system like that, would change everything in the game, so peppy would have options to do that in a comfortable way.

First: Make a score v2, or a score v1 like a button in the mods. Like it is today. The amount of pp could change on the mod you choose (or not).
(You could put the v2 as a main score system and the score v1 like a mod for more pp, like DT or HR. Or make the v2 a mod like EZ making it less worth it. Maybe 90x, or 85x. The same thing you did with mania. When you want to play a 4key song like a 7key song.)

Second: Change everything to the new score V2 system (this is not a good idea because a lot of players would stop playing the game).
If he choose the second option, he would have to change the pp system, puting songs worth it of more pp, less worth it, because everyone can get a good acc in any song if he put a good effort.
Me for example: I can (barely) fc 3.9 stars songs, but there are 5.8 star songs that I can get 98% of accuracy. So, to make it fair, the general quantity of pp would have to decrease.

This is my idea. Thank you for read this whole book that I wrote haha. I'm sorry for this long text.
It could be smaller, I guess. But I didn't master english yet, so I tried to express myself the best way I could.
Lyawi
I didn't read the whole thing but I assume that belongs to Featured Requests. A moderator could move that.
Caput Mortuum
Combo should still give some bonus score, but not so much like the current one. After all, consistency is also a skill and we want to reward skilled players, right?
Topic Starter
Abestado

Eraser wrote:

Combo should still give some bonus score, but not so much like the current one. After all, consistency is also a skill and we want to reward skilled players, right?
Yep
Edgar_Figaro
I'm of the opinion that if the Taiko PP system could be adapted to a scoring system it would be the best. The main idea behind Taiko PP system (ignoring all the SR, song length, OD and such) is that Combo is completely irrelevant...but number of misses matters in addition to Accuracy. This makes it so that missing 3 times is equally detrimental no matter if they are at the beggining, middle, end, or spread throughout the map.

Also 1 or 2 misses won't completely make a score worthless but if you miss too many times then the score is worthless. The current scoring system "kinda" works like the PP system but a single miss takes a huge amount of ACC increase to actually outweigh it in score.

FYI: Taiko PP sytsem also cares about ACC more than # of misses (which makes sense as a miss will lower ACC as well as penalizing for a "miss")
Topic Starter
Abestado

Edgar_Figaro wrote:

I'm of the opinion that if the Taiko PP system could be adapted to a scoring system it would be the best. The main idea behind Taiko PP system (ignoring all the SR, song length, OD and such) is that Combo is completely irrelevant...but number of misses matters in addition to Accuracy. This makes it so that missing 3 times is equally detrimental no matter if they are at the beggining, middle, end, or spread throughout the map.

Also 1 or 2 misses won't completely make a score worthless but if you miss too many times then the score is worthless. The current scoring system "kinda" works like the PP system but a single miss takes a huge amount of ACC increase to actually outweigh it in score.

FYI: Taiko PP sytsem also cares about ACC more than # of misses (which makes sense as a miss will lower ACC as well as penalizing for a "miss")

Well... What I'm proposing to the game is a raw idea. It could change, and adapt for other modes. And if it's not necessary it don't need to be the main system for all of them. After all, they are different games, so the pp/score system could change depending on the mode.
LahiruVIP
combo has influence with the pp since it's about consistency. the higher/better you are the more consitent you are the less misses/sliderbreaks you should get. yes it can be frustarting that you do not FC a song but if you only miss the last couple of notes you are notting a that much high penalty. mid-song it is a big penalty. Also most high playes like cookiezi's freedom dive or rrtyui's big black fox SS are retried multiple times and some players are only 5 years in the game and get close to 10/9/8 year players.

I also can't fc long 4 star songs of like 3min but i can pass them with high acc but they don't give a lot of pp.
however when i play 1 min song i can easily FC them but since you have less strain your consisteny doesn't work here.


I mostly play 1-2 hours a day with sometimes breaks of 1-3 day and still my consistency does build up.

Also higher star song doesn't always give more pp because sometimes then can be easier due the low amount of streams or jumps. scarlet rose is a good example of that ,since it is one of the hardest maps in osu! gives a low amount of pp even with HRHD(around 200pp)



BTW not all osu! player play everyday. most of them take a break and only play osu! for around 3-4.
LunaBintu
It is important that we keep combo in the scoring system for std. one of the biggest issues with removing combo scoring is that a slider break, which breaks combo but doesn’t count as a miss, would be worth the same as an fc. That’s the main issue I can think of currently. Consistency is a very hard skill to work on, and should be heavily rewarded in my opinion.
Vuelo Eluko

Evil_Ocelot wrote:

Consistency is a very hard skill to work on, and should be heavily rewarded in my opinion.
About sums it up. Not to mention how many maps get most of their difficulty from short bursts of very difficult jumps or streams, it makes no sense to reward someone heavily for a map defined in the star rating by those sections if they couldn't do them and just farmed acc from the rest of the map.

Being able to hold combo is a sign of a good player, no matter what you say. That's not to say there aren't skilled players who can't hold a combo, the 'unranked map pros' who just work on achieving hard passes on 8+ star maps and have the reading and speed of a top player but not much else, or that there aren't unskilled players that can hold a combo and have good consistency, take easy/normal/hard ss+scoreboard spinnergod farmers for example.

And of course the above examples are just using what i define as 'skill', people can play however they want and define it how they wish, but I don't think you can just dismiss an entire aspect of the game as 'not requiring skill', especially one as important as consistency.
LahiruVIP

Evil_Ocelot wrote:

Consistency is a very hard skill to work on, and should be heavily rewarded in my opinion.


xxjesus1412fanx wrote:

And of course the above examples are just using what i define as 'skill', people can play however they want and define it how they wish, but I don't think you can just dismiss an entire aspect of the game as 'not requiring skill', especially one as important as consistency.


In every Game consistency is the most required skill to compete at the top. Most skilled players won't make a lot of mistakes since this can result that an other player can punish them for that. example the top player at super smash bros has one the best consistency in the world so other playes can't punished them. In Osu! consistency is only seen during tournament matches because they can't reset in comparison to regular plays where they reset if they miss a note or get a 100.
jesse1412

LahiruVIP wrote:

Evil_Ocelot wrote:

Consistency is a very hard skill to work on, and should be heavily rewarded in my opinion.

xxjesus1412fanx wrote:

And of course the above examples are just using what i define as 'skill', people can play however they want and define it how they wish, but I don't think you can just dismiss an entire aspect of the game as 'not requiring skill', especially one as important as consistency.
In every Game consistency is the most required skill to compete at the top. Most skilled players won't make a lot of mistakes since this can result that an other player can punish them for that. example the top player at super smash bros has one the best consistency in the world so other playes can't punished them. In Osu! consistency is only seen during tournament matches because they can't reset in comparison to regular plays where they reset if they miss a note or get a 100.
No amount of retries will make you fc a map if you're not consistent enough to fc the map...
Full Tablet
A combo system is not a good way to reward consistency.

The importance of accuracy of later notes in the combo is bigger than accuracy in earlier notes. This problem could be prevented while still maintaining a combo system (this is just caused by the specific way combo influences score in this game).

A more important issue is that, with a combo system, the importance of not combo breaking at each note varies according to where it is in the map. A single miss in the middle of the map punishes considerably more than a miss nearer the start or the end of the map, even though there is no justification of why those notes are more important.

The simplest way to solve this problem is changing the combo system by a system that just counts the amount of combo breaks, while maintaining the overall importance of not breaking combo.

Based on scorev1, the combo system would just be replaced by a score multiplier that depends on the amount of combo breaks. The multiplier formula that keeps the importance of an average miss in the map would be:
2 / (2 + Number_of_Combobreaks)
(So a single combo break reduces your score on the play by 33.33%, 2 misses reduce your score by 50%, etc...)

Something similar (but more complex) could be done for creating a system that keeps the importance of not combo breaking of scorev2.
Caput Mortuum

Evil_Ocelot wrote:

It is important that we keep combo in the scoring system for std. one of the biggest issues with removing combo scoring is that a slider break, which breaks combo but doesn’t count as a miss, would be worth the same as an fc. That’s the main issue I can think of currently. Consistency is a very hard skill to work on, and should be heavily rewarded in my opinion.
The thing is, combo weights too much. 90% FC vs 99% 1miss in the middle? fuck you, that 90% is better!
The longer the map, the game becomes more luck based. You can mess up the combo just because your hand itches, loud notification from discord, etc.

Not saying to remove it completely, but just make it weight a little bit less. (even 30% of total score is already big)
abraker
Someone help these people.

Consistency must be valued, yes, however it's being done wrong by using combo mainly due to the middle miss issue. It's better to make it so that a miss anywhere would be equivalent to a miss in the middle so that it at least more fair in terms of skill and a play wouldn't be predetermined until closer to the end. Achieving this is a matter of a steep accuracy curve. It's that simple. One miss would bring the acc to 99.something% and be worth half the max pp. I'm fine with that.

If you are worried that 99.9% on a 10,000 note map is not equivalent to a 99.9% on a 100 note map, pp can be dependent on the number of notes in addition to percent. As an example, it would take 99.98% on a 10k note map to be equivalent to a 99.00% on a 100 note map of the same difficulty.

Full Tablet's miss counting also works just as good.



If peppy decided to go with acc instead of combo to begin with, this dispute wouldn't even exist. Nobody who plays acc wants something as broken as combo. It's only because the std community has only known to how to play for combo that they value it religiously. Their hearts might as well break if the miss -> retry culture vanishes.
jesse1412

Full Tablet wrote:

A combo system is not a good way to reward consistency.

abraker wrote:

Someone help these people.
You couldn't make it more obvious that you're VSRG players if you tried.

Full Tablet wrote:

The importance of accuracy of later notes in the combo is bigger than accuracy in earlier notes. This problem could be prevented while still maintaining a combo system (this is just caused by the specific way combo influences score in this game).
This is the one true problem with the current system and it's fixed by rewarding score on a map wide basis rather than a per note basis. Keep a log of all of the combos throughout a map, then take the average value of combo across all notes and multiply it by accuracy. Formula wise you would just sum 1...n of each combo streak to get the value of the combo streak, then take the average values of those sums to find average value of combo across the map. Multiply the average combo value by accuracy*100 and you have an improved scoring system. This system would work effectively the same as the current one but all 300/100/50s would be worth the same amount regardless of combo.

abraker wrote:

Consistency must be valued, yes, however it's being done wrong by using combo mainly due to the middle miss issue. It's better to make it so that a miss anywhere would be equivalent to a miss in the middle so that it at least more fair in terms of skill and a play wouldn't be predetermined until closer to the end.

Full Tablet wrote:

A more important issue is that, with a combo system, the importance of not combo breaking at each note varies according to where it is in the map. A single miss in the middle of the map punishes considerably more than a miss nearer the start or the end of the map, even though there is no justification of why those notes are more important.
Completely untrue. The reason that a miss in the middle hurts more than a miss near the start/end is because it takes more consistency to get a combo across 90% of the map than 50%. It makes sense that where you missed matters when taking consistency into account.

Full Tablet wrote:

The simplest way to solve this problem is changing the combo system by a system that just counts the amount of combo breaks, while maintaining the overall importance of not breaking combo.
Even if I agreed with you that's not a good solution. This system wouldn't consider slider breaks as misses.

But I don't agree with you. With your system, someone who misses the last note in a map has their score COMPLETELY removed from the top rankings if everyone else fcs. Someone who has 99999/100000 combo and 99.999% accuracy would be beaten by someone with an 80% acc fc. Your proposed system would be worse from both of our perspectives.

Eraser wrote:

The thing is, combo weights too much. 90% FC vs 99% 1miss in the middle? fuck you, that 90% is better!
The longer the map, the game becomes more luck based. You can mess up the combo just because your hand itches, loud notification from discord, etc.

Not saying to remove it completely, but just make it weight a little bit less. (even 30% of total score is already big)
The issue is that there ARE cases where a 90% fc would be more impressive than a 99% score that missed in the middle. A very obvious example is a TV size jump map with a huge ending spike; accuracy during the first 90% of the map shouldn't be rewarded anyway because those sections are hilariously easy, and the person who achieved 99% could have done so while missing every note in the hardest pattern.

I understand your point that accuracy should matter and I agree with you, but it's not a clear cut argument like you lay it out to be.

abraker wrote:

Achieving this is a matter of a steep accuracy curve. It's that simple. One miss would bring the acc to 99.something% and be worth half the max pp. I'm fine with that.
Completely ridiculous. First of all you're discussing the reward of accuracy for the pp system while we're talking about the scoring system. Secondly, you're essentially extending the "problem" to be even worse. Rather than a miss in the middle being crippling, you want a miss anywhere to be crippling? That goes completely against your own argument that combo is valued too much over acc.

abraker wrote:

If peppy decided to go with acc instead of combo to begin with, this dispute wouldn't even exist. Nobody who plays acc wants something as broken as combo. It's only because the std community has only known to how to play for combo that they value it religiously. Their hearts might as well break if the miss -> retry culture vanishes.
If Konami decided to go with combo instead of acc to begin with, this dispute wouldn't even exist. Nobody who plays combo wants something as broken as acc. It's only because the VSRG community has only known to how to play for acc that they value it religiously. Their hearts might as well break if the 100s -> retry culture vanishes.

News flash, retrying for acc and retrying for combo are BOTH difficult and frustrating. Standard osu! players generally strive to achieve both while VSRG players tend to only value accuracy. Both of the arguments above are absolutely retarded. Combo and accuracy should BOTH matter. The only deluded one here is you for the exact reasons you put forward, the VSRG community has an accuracy only mindset engraved by konami.

You won't be in the leaderboards for having only good accuracy, but you might for having only a good combo. That is the only point that you can and should put forward. The issue is there's no proper way to fix this without a per-map scoring system. It's impossible to create a catch all scoring system because the difficulty of getting high acc vs high combo is different for every map.

Just in case anyone forgot, pp is a thing and it essentially can create per-map scoring systems. The only solution I've ever come across for all these issues is to use a system like pp as the scoring system. Ideally time would be allocated to creating an improved pp system rather than another shitty scoring system. It's sad that pp STILL isn't as robust as tp was five years ago. Until aim/speed/acc are split for each score, we won't be seeing any pp improvements on the level of tp imo.
Full Tablet

jesse1412 wrote:

You couldn't make it more obvious that you're VSRG players if you tried.
I have played both standard and VSRG games a considerable amount of time, I have had time to think this matter from several perspectives.

jesse1412 wrote:

Completely untrue. The reason that a miss in the middle hurts more than a miss near the start/end is because it takes more consistency to get a combo across 90% of the map than 50%. It makes sense that where you missed matters when taking consistency into account.
If a player gets a single miss in the middle of a map, and another player gets a single miss 90% through the map, I don't see a reason to judge that one player is more consistent than the other.

One possible reason might be that players might have a higher psychological stress when their combo is higher, thus each subsequent note in the combo might be harder to hit.

First of all, I am not convinced that higher stress does really have an overall negative impact on performance. This could be very person-dependent, with many players actually playing better because they are more focused; or there would be people that aren't affected by this at all. This is something that would need to be studied carefully in order to determine.

Also, even if having higher combo inherently made keeping combo harder (all of this assuming it is not known which parts of the map have harder patterns, something that matters considerably more than the hypothetical effect combo has in difficulty), it is not really straightforward to judge that the player who missed in the harder part of the map is more consistent than the player that missed in the easier part. One could create many arbitrary systems where players are punished more because they failed in the most basic elements of the map, or the opposite, or something in-between; there are no strong arguments that make you choose one system over the others.

No matter what system you choose, as the probability a certain player hits each note in a map approaches 100% (or 0%), the less it should matter whether the player missed in the harder or easier part (as a consequence of Le Cam's theorem); in practice, most players have a chance of hitting each individual note close to 100% (they tend to hit several hundreds of notes and just miss a few); so the inherent subjectivity in the system chosen should be only significant in cases where there are very intense peaks of difficulty in a map (which happens very often because of the osu! standard mapping meta).

jesse1412 wrote:

Full Tablet wrote:

The simplest way to solve this problem is changing the combo system by a system that just counts the amount of combo breaks, while maintaining the overall importance of not breaking combo.
Even if I agreed with you that's not a good solution. This system wouldn't consider slider breaks as misses.

But I don't agree with you. With your system, someone who misses the last note in a map has their score COMPLETELY removed from the top rankings if everyone else fcs. Someone who has 99999/100000 combo and 99.999% accuracy would be beaten by someone with an 80% acc fc. Your proposed system would be worse from both of our perspectives.
The system could count both slider breaks and misses as combo breaks. Scores are calculated during or just after gameplay, so all the required information is available at that moment.

In my perspective, I don't really care whether or not avoiding misses is important or not (I have a stronger opinion about this matter in VRGS, since they don't have an aim aspect); the system I proposed solves the problem of varying importance of each note based on their position in the map, while still keeping the importance of not missing.

The formula 2/(2+Number_of_Combobreaks) associated with scorev1 (quadratic growth of score with combo length, with an insignificant linear part) is based on the average relative penalty certain amount of combobreaks have, when there is not information of where the misses are, and assuming each note in the map has an equal chance of being missed.

Your example is an edge case that shows the situation rather well. If a player was somehow destined to miss exactly once in a certain extremely long map, he would be very lucky if that miss was just at the very last note with a combo system, since the penalty would be relatively very small; in that situation, without any further previous information, the penalty of missing once in that map would be 33.33% in average in scorev1 (which would make him lose against the low accuracy player who gets a FC, something very few people agree with).
Vuelo Eluko

Full Tablet wrote:

If a player gets a single miss in the middle of a map, and another player gets a single miss 90% through the map, I don't see a reason to judge that one player is more consistent than the other.
Because the the player that missed near the end held a combo for longer.. which takes more consistency.
LahiruVIP

xxjesus1412fanx wrote:

Full Tablet wrote:

If a player gets a single miss in the middle of a map, and another player gets a single miss 90% through the map, I don't see a reason to judge that one player is more consistent than the other.
Because the the player that missed near the end held a combo for longer.. which takes more consistency.
It also takes more stamina since your fingers will get tired when you are playing for a long time.
Full Tablet

LahiruVIP wrote:

It also takes more stamina since your fingers will get tired when you are playing for a long time.
Missing a note doesn't make your fingers become less tired.

xxjesus1412fanx wrote:

Because the the player that missed near the end held a combo for longer.. which takes more consistency.
Your definition of consistency seems to be "holding a combo for a long time". If that's the case, this is a circular argument, otherwise, I don't get what you are getting at.
jesse1412

Full Tablet wrote:

xxjesus1412fanx wrote:

Because the the player that missed near the end held a combo for longer.. which takes more consistency.
Your definition of consistency seems to be "holding a combo for a long time". If that's the case, this is a circular argument, otherwise, I don't get what you are getting at.
Keeping a combo shows aiming consistency, a key component of osu! gameplay. I don't understand what other method there is to measure aim consistency. Your view seems to not value the aim aspect of osu!.
show more
Please sign in to reply.

New reply