forum

osu! scoring system.

posted
Total Posts
27
Topic Starter
Horolynn
Why does it rely on combo so much?
The one thing that bugs me the most about it atm is how combo means absolutely everything as far as score is concerned, it really makes no sense to me.
Say, you've got two players playing the same map, they've slipped one slider beginning each, let's say they're both at 99.8% accuracy total.
But, the first player slipped on a slider in a middle map, but for comparison's sake, the second player did so at the end of the map, but the exact same slider, since the pattern from the middle of the map was reused. They both broke their combo, yes, but since player 1 broke it in the middle of the map, he's nowhere near player 2's score. Keep in mind the slider they missed was exactly the same, just at a different point of the map. They did equally well as far as I'm concerned, and there is no reason to punish the first player for making the mistake at a different time than the second player.
Right now, I don't even feel like continuing some maps when I break combo, it's really frustrating that I can't reach my other scores simply because I made a mistake at a different point in the map.
I want to note, I'm not trying to start a revolution here. I'd like other people's point of view on that topic, and how do they feel about how it is right now.
GoldenWolf

Draxuss wrote:

Why does it rely on combo so much?
Because aiming is an important part of the game.

The second player deserves more points because he could keep a combo through the whole map, which is harder than keeping a combo for half of the map.
It can be harder because the second half of the map has more jumps and all, or because you can get more nervous the closer you are from FCing, etc.
Almost
I find the combo system a way of showing whether or not you are able to play the map or not. Being able to FC a map almost guarantees that you are able to play every single pattern accurately. If it weren't like this, then there would be plenty of players neglecting their aim and reading and just developing accuracy to get higher ranks. Though it is possible that you can get lucky, but this generally doesn't happen often.
Wishy
Score system is bad to a point where a lousy 84% acc play can beat a x1 miss 99.99% play.

It would be really tough to make a proper score system for this game since there are maps where acc is easy but it's also easy to miss, and of course the other way around is possible too. So yeah.
VoidnOwO
Combo works on maps like BD's diff on 'he has no mittens' that are easy, but there's that one part that 99% players can't FC.
Combo is shit on 8 minute dragonforce maps where, like Wishy said, 80% acc play could beat a 99,99% play.


Theoretically, wouldn't Taiko system work quite well in osu! too? (Combo bonus capping at X combo). Combo would make a difference, but accuracy would matter the most.
Topic Starter
Horolynn

GoldenWolf wrote:

The second player deserves more points because he could keep a combo through the whole map, which is harder than keeping a combo for half of the map.
It can be harder because the second half of the map has more jumps and all,
or because you can get more nervous the closer you are from FCing, etc.
They missed the same thing, they kept their combo throughout everything else evenly.
Okay, that makes sense. But osu! should not be a test of one's psychological strength, or at least should not be mainly that.

Almost wrote:

I find the combo system a way of showing whether or not you are able to play the map or not. Being able to FC a map almost guarantees that you are able to play every single pattern accurately. If it weren't like this, then there would be plenty of players neglecting their aim and reading and just developing accuracy to get higher ranks. Though it is possible that you can get lucky, but this generally doesn't happen often.
No, 100% accuracy would still be better than 98% accuracy. If there would be players neglecting their aim and reading, they'd still be worse than players who would not do that. Being able to FC the map shows that you can play it all the way through, yeah. But really, is 90% FC better than 97% 1 miss play?

Wishy wrote:

Score system is bad to a point where a lousy 84% acc play can beat a x1 miss 99.99% play.
That's exactly what I'm getting at. I understand that FC is okay as it is, since you've just played through a map perfectly, aside from accuracy unless you've got 100% too. But situations like this one are what I can't understand.

BRBP wrote:

Theoretically, wouldn't Taiko system work quite well in osu! too? (Combo bonus capping at X combo). Combo would make a difference, but accuracy would matter the most.
I think that actually just simply leaving it as it is, without the combo multiplier, would be fine. You'd still have the "color combos" that earn you small bonuses at the end of each color depending how well did you do.
GoldenWolf

Draxuss wrote:

Wishy wrote:

Score system is bad to a point where a lousy 84% acc play can beat a x1 miss 99.99% play.
That's exactly what I'm getting at. I understand that FC is okay as it is, since you've just played through a map perfectly, aside from accuracy unless you've got 100% too. But situations like this one are what I can't understand.
Situations like this one are EXTREME, and if a player can do 99.5+ % then he can FC the map if he cares enough to retry, the 84% FC play was most probably pure luck.

Draxuss wrote:

Okay, that makes sense. But osu! should not be a test of one's psychological strength, or at least should not be mainly that.
Keeping a Full Combo is actually more of that lol.
If one can do at least 99% on a map then he can FC it (unless it has ridiculous jump/pattern that almost no one can FC or w/ever), the hardest part of FCing is to stay focused through the whole map.
Topic Starter
Horolynn

GoldenWolf wrote:

if a player can do 99.5+ % then he can FC the map if he cares enough to retry
Why does he have to retry just because of that small dumb mistake? If he'd miss at the end of the map he wouldn't have to, that is my point.
People that can FC the maps would still be in a good standing if they can maintain a high accuracy. If they can't, what does it matter that they didn't miss?
To me, getting 50 x5 and say 100x 10 is way further away from perfect than getting miss x1.


GoldenWolf wrote:

the hardest part of FCing is to stay focused through the whole map.
If you wouldn't have to worry about FCing the map, you'd still have to worry about your accuracy. Do you think that getting a high accuracy doesn't actually require you to stay focused?
thelewa
being able to get good accuracy at a map doesn't mean you'll be able to FC it with enough tries

I've had HDHR maps where I constantly miss at a hard part and SS the map otherwise, and any amount of retries in the map isn't going to make me better at that hard part so I'll be stuck at getting 1 miss SS runs on it until I get better at the game by playing other maps

besides, some people have bad consistency in terms of aim but not in terms of accuracy, so they will find it exceedingly difficult to FC jumpy maps, but they will never get 100's in them.
Topic Starter
Horolynn

thelewa wrote:

besides, some people have bad consistency in terms of aim but not in terms of accuracy, so they will find it exceedingly difficult to FC jumpy maps, but they will never get 100's in them.
And in my opinion, (inb4 no one cares about your opinion, I realize that might be the case) accuracy is way more important than hitting every single note.
GoldenWolf
But we're in a game where one of the most important part is the aim, hitting every single note perfectly is a part of osu!, that's the main difference with other rhythm games.
JappyBabes
If you want a mode which is almost entirely based on accuracy, play taiko or mania. Aim is an interesting mechanic to be used in a rhythm game as it's not used in them (at least to my knowledge). Standard isn't just a rhythm game, I've always seen it as primarily a score attack game first and rhythm second. If you're someone who is consistently retrying yet not FCing, I wouldn't even consider calling you as good of a player as someone who could FC it even after retrying and with your higher accuracy in mind. You're simply just not on that level. And getting a 50 or a 100 is better than a miss, objectively speaking. It provides more accuracy than a miss and it shows you were actually able to aim properly. A predominantly accuracy based system for a mode which involves aim could also be heavily abused, you could just miss one or two of the notes in hard patterns in the map and play everything else well and suddenly you've got one of the best plays on the map even though you took out the hardest parts.
boat
In what scenario does score actually even matter though? It only matters as much as you want it to.
VoidnOwO

Draxuss wrote:

BRBP wrote:

Theoretically, wouldn't Taiko system work quite well in osu! too? (Combo bonus capping at X combo). Combo would make a difference, but accuracy would matter the most.
I think that actually just simply leaving it as it is, without the combo multiplier, would be fine. You'd still have the "color combos" that earn you small bonuses at the end of each color depending how well did you do.
You don't think that a 95% FC in BD's diff here should have a higher score than a 96% with a combo break?
Almost

JappyBabes wrote:

A predominantly accuracy based system for a mode which involves aim could also be heavily abused, you could just miss one or two of the notes in hard patterns in the map and play everything else well and suddenly you've got one of the best plays on the map even though you took out the hardest parts.
Exactly this. I find someone who can consistently FC a map better than someone who has to luck out to FC it.
Full Tablet
Well, I think that here is 2 separate issues with the scoring system mentioned here:

A) The order of the misses/sliderbreaks/50s/100s/300s matters, specially with misses and slider breaks (since they set the combo to 0). This leads to the case where 2 players get very different scores even though they had the same amount of misses/sliderbreaks/100s/300s/50s.

IMO the scoring should either not consider the order of the judgments obtained during the play at all, or weight difficult parts higher (though this is not really necessary; even if a player "a" plays difficult parts better than player "b", if player "a" plays worse than player "b" in the easy parts then it's not really clear who is the best player, and at that point it becomes very subjective).

That way the case Draxuss mentioned in his initial post doesn't happen. As for the idea that keeping a higher combo value though the map should give more score, since it is harder, I think it heavily has to do with the map that is being played (it can vary a lot between different maps): in some maps it can be easy to not combo break early in the song, while in others it is easier to keep the combo in the latter parts. The current combo system would be more adequate if keeping high combo showed more skill in the map consistently (when comparing plays with similar accuracy).

B) Combo breaks affect score a lot (so, most of the time, a play "a" that has less combo breaks will score higher than another play "b" with more combo breaks; even if the play "a" has much worse accuracy than a play "b").

IMO the way the misses and slider-breaks affect the score should be tweaked so it doesn't rely so much on Aim, but misses should still have a considerable effect on score (so aim can't be disregarded completely).

For example: in a long map.
Play A: 99.5%acc FC.
Play B: 99.4%acc with one miss in the middle.
Play B": 99.4%acc with one miss in the beginning.
Play C: 90%acc FC.
Play D: 99%acc FC.

With current scoring the scores would look like this:
A: ~ 20million
B: ~ 10million
B": ~ 19.9million
C: ~ 18million
D: ~ 19.9million

While I think something like this would be more reasonable:
A: ~ 20million
B: ~ 19.2million
B": ~ 19.2million
C: ~ 16.3million
D: ~ 19.4million
Loves
It's part of the skill to remain steady throughout the whole map.
peppy
why does a game work the way it does, rather than like this other game does? why are rules this way? let's reinvent the rules! let's change everything so it can be how *I* want to play it. the way osu! does something is silly because of biased opinion.
Topic Starter
Horolynn

peppy wrote:

why does a game work the way it does, rather than like this other game does? why are rules this way? let's reinvent the rules! let's change everything so it can be how *I* want to play it. the way osu! does something is silly because of biased opinion.
In my original post I stated that I'm not here to push my "great ideas" anywhere, I just want to know how do people feel about the way it is, and if they can show me advantages to the scoring system that is currently in place.
peppy
if you search back, this discussion thread is made about once a week. you might want to go read previous ones. i have commented many times, including my opinion on it and how i fixed it in osu!stream etc.
TheVileOne
The system favors consistency rather than accuracy. Personally I believe the biggest strength in such a system is score variation. There would be a ton more similar scores, because combo breaks don't vastly change how the scores are calculated. Combo would be rendered practically meaningless. I don't agree with changing combo from something important to something almost devoid of meaning.
Topic Starter
Horolynn

peppy wrote:

if you search back, this discussion thread is made about once a week. you might want to go read previous ones. i have commented many times, including my opinion on it and how i fixed it in osu!stream etc.
Alright, I should've searched first. My bad.
Yarissa
Although there may be threads like this quite often I find no harm in sharing eachother's opinions. If someone didn't post these threads I'm sure a handful of new players somewhere would be wondering the SAME sorts of things.

On topic: I think accuracy is a more important element in rhythm games. It judges your ability to time notes which is kind of what a rhythm game is about. Comboing, on the other hand, is a really rewarding feature which makes the game really fun and encourages you to aim better whenever you play. If you compare aim and accuracy side by side 9 times out of 10 most people will say "aim is harder." Which is probably why comboing maps yields the most score. And although SOME misses may not make sense or the scoring system may not always consistently measure your ability to play, the combo system that is currently in place is probably better than any ideas most of the players here can innovate. And if you can, good luck coding it.
RaneFire
Well, I'll share my thoughts again then:

The thing about rhythm games is that "rhythm" is a general term and can mean many things, and rightfully so. It is just "hitting buttons to the music." But because of this people seem to think that osu! is like any other rhythm game, considering this definition, but it is not.

MOST rhythm games have a far more complex accuracy system than osu!standard, and require more thought and controlled response to patterns presented, but are solely based on your accuracy with regard to reading those patterns. In this game, to make up for the difference in the "response complexity" to rhythm, your ability to read patterns is split between aiming and accuracy. Your ability to play this game does depend on how consistently you can aim, and the only way to do that is with a combo system. i.e. connected hits.

If accuracy was all that mattered the aim component would be nothing more than a gimmick, which if you wanted, you could skip the hardest jumps in the map, focus on hitting the following notes after very accurately, and still make a top50. This is not just about getting 1x miss anymore, because you tried to hit that 1x miss, but now you have the option not to. Trying to hit very hard jump patterns does adversely affect your accuracy, because it requires more focus on aiming.

On the contrary if osu! were to be fully focused around rhythm accuracy, it would require a revamp of the scoring system and hit error leniency in its entirety, and getting the "high accuracy scores" you are deluded into believing are good now, would not be, because the mapping meta would have also changed long ago to make getting high accuracy harder. As far as "hitting your buttons" goes, mapped patterns these days are really easy, there are so few fingerfuck maps and they are terribly unpopular. The focus is mainly on making it even easier to play now because of "flow mapping," which basically implies that you can predict the map.
Tidek
The main problem of scoring in osu!:

- Getting 100 on beginning of song costs you 200-2000points
- Getting 100 on the end of song costs you 10k or more points

It sucks, especially on marathon maps.
Topic Starter
Horolynn

RaneFire wrote:

Well, I'll share my thoughts again then:

The thing about rhythm games is that "rhythm" is a general term and can mean many things, and rightfully so. It is just "hitting buttons to the music." But because of this people seem to think that osu! is like any other rhythm game, considering this definition, but it is not.

MOST rhythm games have a far more complex accuracy system than osu!standard, and require more thought and controlled response to patterns presented, but are solely based on your accuracy with regard to reading those patterns. In this game, to make up for the difference in the "response complexity" to rhythm, your ability to read patterns is split between aiming and accuracy. Your ability to play this game does depend on how consistently you can aim, and the only way to do that is with a combo system. i.e. connected hits.

If accuracy was all that mattered the aim component would be nothing more than a gimmick, which if you wanted, you could skip the hardest jumps in the map, focus on hitting the following notes after very accurately, and still make a top50. This is not just about getting 1x miss anymore, because you tried to hit that 1x miss, but now you have the option not to. Trying to hit very hard jump patterns does adversely affect your accuracy, because it requires more focus on aiming.

On the contrary if osu! were to be fully focused around rhythm accuracy, it would require a revamp of the scoring system and hit error leniency in its entirety, and getting the "high accuracy scores" you are deluded into believing are good now, would not be, because the mapping meta would have also changed long ago to make getting high accuracy harder. As far as "hitting your buttons" goes, mapped patterns these days are really easy, there are so few fingerfuck maps and they are terribly unpopular. The focus is mainly on making it even easier to play now because of "flow mapping," which basically implies that you can predict the map.
Okay, I can understand why is it the way it is right now. Thanks for all the input on the matter, everyone.
Reyvateil

RaneFire wrote:

snip
And the fact that even if there were some scoring system which escalated with combo similar to Taiko and the harder accuracy requirement as you suggested, kinda balancing both the combo (aiming) importance with the accuracy importance, the changes from the meta from actually implementing it would mean the whole current ranking system would have to be wiped to much displeasure of some, which wouldn't be very beneficial to the community since there would be people rage quitting for sure.

I kinda have to agree with peppy at this, if some changes are going to be made concerning this it might be better to just revamp the whole game into something new like Osu!Stream, not only adjusting the scoring system, but also mods to fit the new changes, and some new stuff like adding the hold notes from Osu!Stream...
Please sign in to reply.

New reply