1. osu! forums
  2. osu!
  3. Gameplay & Rankings
posted
Hi all osu! players! I'll go straight to the point. Everybody knows that combos play a huge part when calculating scores, as it's said in the osu! wiki:

Most of the time, combo plays the major part in the scoring system since it serves as an multiplier of the score.
I really insist that this is totally wrong and disagree this formula. Do you really think that playing a full combo including less or more 100's and 50's would be better than a play of only 300's and one mistake in the middle of the song? Scoring would give the former better result just because of the so-called combing. I have got some S ranked scores which are, for the above-mentioned reason, less score than my previously scored C's. I have noticed that, practically always, the better combo means better score and vice versa.

I think it shouldn't be like this. This score formula really distorts the reality.

I could use very rough langugae at this point, but I say it neutrally: losing your combo reduces your score much more than it should. It is really annoying. Personally I always use accuracy as win condition in multiplayer.

This post in a nutcase: Combos should not count as much as they do at the moment!
posted
It is how it is, and it won't be changed.
posted
posted

GenoClysm wrote:

Wrong section buddy, move to gameplay and rankings.
or suggestions or something

As for this suggestion, this shouldn't be changed and will not be changed. rhythm games are almost completely based off of combos, that's how it's been and will always be.
Aswell, overall skill in rhythm games is more consistency than anything, and the only way that can be told is through combos.
Combos also act as a multiplier that makes for much more score diversity, making perfection more perfect and seperating unique scores. If you haven't noticed, there are easy and normal maps with 50+ people with literally the exact same score, all of them getting SS's. If combos didn't completley make score in harder maps, we'd have a similar situation likely. Everyone with 1 miss will have the exact same score, everyone with 2 100's will have the exact same score, etc etc.

What you are asking for is impossible. Rhythm games and competition in them are (almost)COMPLETELY structured around combos, and without them being almost entirely the weight, rankings would be in chunks of people, not individuals lol
Good points! You said that people would get exactly same scores. I dind't say that combos shouldn't count at all, just that they should count much less. FOr example in Guitar Hero, Rock Band and other quality rythm games, the maximum multiplier is 4x. That's way better. Well, and don't forget the spinning extra points.

YayMii wrote:

peppy has been aware of this problem for a while already. The score system can not be changed though, as such change will result in millions upon millions of scores getting wiped. This is one of the reasons why the pp system weighed other factors more heavily than score alone.
What you said that scores would be wiped, that's not really true. A rpgram could calculate all the scorse again based on the replays and then re-order the lists. I know it would be a shock a first, but it would work.


I am not assuming that this kind of major change would really happen, but at least to see how many people agree with me.
posted
One half of the game is getting accuracy while the other half is about aim. Combo is the aim portion of the game so there is nothing wrong with the formula.
posted

RaneFire wrote:

http://osu.ppy.sh/forum/t/157821
http://osu.ppy.sh/forum/t/149146

Read this part: http://osu.ppy.sh/forum/p/2604353
I read all of those. They are logical, and perhaps I have to re-think. But still, that the multiplier grows this fast as it grows in osu!, is not very smart. As I said, for example in Guitar Hero and Rock Band maximum multipliers are the class of 4. And that's somthing osu! could learn from.
posted

Pelaaja_X wrote:

I read all of those. They are logical, and perhaps I have to re-think. But still, that the multiplier grows this fast as it grows in osu!, is not very smart. As I said, for example in Guitar Hero and Rock Band maximum multipliers are the class of 4. And that's somthing osu! could learn from.
It could be better. It still wouldn't be perfect.

Here's the thread OP was quoting from: http://osu.ppy.sh/forum/t/181312
Don't make a new thread here. In this quote, he is asking the moderators to move the thread. There was no need to duplicate. They'll probably merge the two.

GenoClysm wrote:

Wrong section buddy, move to gameplay and rankings.
or suggestions or something
posted
peppy thinks the game would be better this way (but he can't change it now, it would upset too many people) and I agree, accuracy is a better measurement than combo in my opinion.
posted

Rewben2 wrote:

peppy thinks the game would be better this way (but he can't change it now, it would upset too many people) and I agree, accuracy is a better measurement than combo in my opinion.
Exactly. Thanks for your agreement.
posted

RaneFire wrote:

http://osu.ppy.sh/forum/t/157821
http://osu.ppy.sh/forum/t/149146

Read this part: http://osu.ppy.sh/forum/p/2604353
I think this brings up a good point, perhaps combo is a better system for rankings. But I think accuracy would be better for things like the upcoming ladder system or the OWC, because anyone can lose concentration for a moment and miss while someone else, who may be doing much worse in terms of accuracy, would fc and win.
posted

Rewben2 wrote:

peppy thinks the game would be better this way (but he can't change it now, it would upset too many people) and I agree, accuracy is a better measurement than combo in my opinion.
I don't get it. osu! is about aim, not accuracy, why would you make it the main measurement?
posted

GoldenWolf wrote:

Rewben2 wrote:

peppy thinks the game would be better this way (but he can't change it now, it would upset too many people) and I agree, accuracy is a better measurement than combo in my opinion.
I don't get it. osu! is about aim, not accuracy, why would you make it the main measurement?
I'd like to think it's about both but just leaning a bit more towards aim.
posted

GoldenWolf wrote:

Rewben2 wrote:

peppy thinks the game would be better this way (but he can't change it now, it would upset too many people) and I agree, accuracy is a better measurement than combo in my opinion.
I don't get it. osu! is about aim, not accuracy, why would you make it the main measurement?
I explained myself - I believe rankings should stay the same but things like OWC should rely on accuracy if you're trying to determine who the better team/player is. Anyone can randomly drop and be consistent throughout the rest of the song - While someone else who is constantly getting 100's may fc with an unimpressive accuracy and win. I think it's just a more accurate way of determining who is a better player opposed to combo. Perhaps misses could be more of a detriment to accuracy than it is now if there was a system that was accuracy based, to prevent people doing what was listed in the thread above (Ignoring hard parts of the song and just doing well on the rest).

But rankings should stay the same because then people can't ignore parts of the song at all. A lot of songs are just who can get the higher % fc, though.
posted

Rewben2 wrote:

RaneFire wrote:

http://osu.ppy.sh/forum/t/157821
http://osu.ppy.sh/forum/t/149146

Read this part: http://osu.ppy.sh/forum/p/2604353
I think this brings up a good point, perhaps combo is a better system for rankings. But I think accuracy would be better for things like the upcoming ladder system or the OWC, because anyone can lose concentration for a moment and miss while someone else, who may be doing much worse in terms of accuracy, would fc and win.
About the point that the combo system gives variation to scores: it is possible to make a combo system at the same time that serves the same purpose, but without making the score so dependent on it:
For example: The current score given per hit is
Hit Value * (1 + Combo multiplier * Difficulty multiplier * Mod multiplier / 25 )
An alternate could be this (I don't say this would be a good formula, it is just a simple one to illustrate the point):
Hit Value * (1 + LOG10(1 + Combo multiplier * Difficulty multiplier * Mod multiplier / 25 ))

With old formula (scores are relative to each other): If anything that is not a combo break is a 300.
1000 combo full combo: ~20million
900combo + 99 combo: ~16million
500combo+ 499 combo: ~10million
300combo+400combo+298combo: ~6.8million
Alternate Formula:
1000 combo full combo: ~0.84million
900combo + 99 combo: ~0.82million
500combo+ 499 combo: ~0.78million
400combo+300combo+298combo: ~0.73million

With combo effect decreased then accuracy becomes more important (while combo still has enough effect to generate variation).

About the fact people can beat other players by ignoring circles in hard to aim maps, then there can be some changes in what misses do. A very simple change would be changing misses so they reduce score by 6000 and break combo instead of just breaking combo.
posted
Isn't not missing/keeping a combo an integral part of the game? (aim)
Being consistent in not missing is a skill in itself.
posted

Full Tablet wrote:

I think this brings up a good point, perhaps combo is a better system for rankings. But I think accuracy would be better for things like the upcoming ladder system or the OWC, because anyone can lose concentration for a moment and miss while someone else, who may be doing much worse in terms of accuracy, would fc and win.
About the point that the combo system gives variation to scores: it is possible to make a combo system at the same time that serves the same purpose, but without making the score so dependent on it:
For example: The current score given per hit is
Hit Value * (1 + Combo multiplier * Difficulty multiplier * Mod multiplier / 25 )
An alternate could be this (I don't say this would be a good formula, it is just a simple one to illustrate the point):
Hit Value * (1 + LOG10(1 + Combo multiplier * Difficulty multiplier * Mod multiplier / 25 ))

With old formula (scores are relative to each other): If anything that is not a combo break is a 300.
1000 combo full combo: ~20million
900combo + 99 combo: ~16million
500combo+ 499 combo: ~10million
300combo+400combo+298combo: ~6.8million
Alternate Formula:
1000 combo full combo: ~0.84million
900combo + 99 combo: ~0.82million
500combo+ 499 combo: ~0.78million
400combo+300combo+298combo: ~0.73million

With combo effect decreased then accuracy becomes more important (while combo still has enough effect to generate variation).

About the fact people can beat other players by ignoring circles in hard to aim maps, then there can be some changes in what misses do. A very simple change would be changing misses so they reduce score by 6000 and break combo instead of just breaking combo.[/quote]

Yeah, in another system a miss could just be more penalising than it is now but not so much that it makes or breaks a win/loss.

Philantropist wrote:

Isn't not missing/keeping a combo an integral part of the game? (aim)
Being consistent in not missing is a skill in itself.
I'm aware of that and agree, which is why I think rankings should rely on it. But for tournaments owc etc etc., I think accuracy is better than combo. I think most people would agree that a 99.5% with a miss is more impressive than a 95% fc.
posted

Rewben2 wrote:

I'm aware of that and agree, which is why I think rankings should rely on it. But for tournaments owc etc etc., I think accuracy is better than combo. I think most people would agree that a 99.5% with a miss is more impressive than a 95% fc.
Your point on scoring for tournaments is debatable really. The players selected would be quite different in an accuracy based tournament since aim type players would lose out to an accuracy based player unless you count really jumpy maps. Also, a low skilled player can beat higher skilled player in this type of match in the IC I managed to beat Shott in an accuracy battle even though I have a lot worse aim and speed then him and would lose to any other type of match.
posted

Rewben2 wrote:

I think most people would agree that a 99.5% with a miss is more impressive than a 95% fc.
It totally depends on the map.
posted

GoldenWolf wrote:

Rewben2 wrote:

I think most people would agree that a 99.5% with a miss is more impressive than a 95% fc.
It totally depends on the map.
I guess so for a map like airman with those jumps near the start I don't even know if that's a good example at all but I think generally speaking, if the map is pretty similar difficulty throughout then the higher accuracy would be more impressive. 5% away from an SS is a huge difference compared to .5% away from an SS.

What is "more impressive" is opinion based, but that's my take on it.
posted

Almost wrote:

Rewben2 wrote:

I'm aware of that and agree, which is why I think rankings should rely on it. But for tournaments owc etc etc., I think accuracy is better than combo. I think most people would agree that a 99.5% with a miss is more impressive than a 95% fc.
Your point on scoring for tournaments is debatable really. The players selected would be quite different in an accuracy based tournament since aim type players would lose out to an accuracy based player unless you count really jumpy maps. Also, a low skilled player can beat higher skilled player in this type of match in the IC I managed to beat Shott in an accuracy battle even though I have a lot worse aim and speed then him and would lose to any other type of match.
Yeah, creating a system for a ladder in a rhythm game that is actually accurate and will determine the better player each time is incredibly hard because of inconsistencies. Don't you think the whole "low skilled player can beat higher skilled player" is more present in a combo-based system where a single mistake can determine the game, instead of relying on how well you do throughout the map?
show more
Please sign in to reply.