TakuMii wrote:

Just a question to all the people opposing slider accuracy: What if good accuracy was implemented as a small score bonus rather than penalizing bad accuracy?

I know it's effectively the same thing, but I feel like psychologically it would be better for the game to provide incentive for going the extra mile, rather than punishing players for playing the sliders the way they always have. (not to mention it could be balanced and adjusted more compared to simply slapping players with a 100 or a 50).

Yeah, I'm aware that many maps aren't quite designed with slider accuracy in mind, but just remember that slider accuracy implementation has been an idea for the developers even as far back as 2011. There's several reasons why it hasn't happened yet, but with all of the evolution to the game happening behind the scenes recently, I think it's the perfect time for the developers to finally try to make things work. I'm all for adding more depth to the game, as long as it doesn't cause more frustration with the people who prefer the status quo.

Railey2 wrote:

Missing out on that bonus when not hitting them correctly will feel like a penalty for players that care about acc.

TakuMii wrote:

Railey2 wrote:

Missing out on that bonus when not hitting them correctly will feel like a penalty for players that care about acc.

What I meant was to keep accuracy calculation as it is now, and give only score bonuses to accurate slider hits, similar to how rainbow 300s differ from standard 300s in mania (as in, you can still get 100% accuracy/SS without the bonuses, but with a lower score). I know a lot of other rhythm games take an approach like this too, so it might be a compromise worth thinking about.

...I'm just trying to come up with ways that they could implement slider accuracy without making it frustrating for people. And I highly doubt they'd want to cause a divide in the community by forcing the mappers to make the decision for them.

Endaris wrote:

Problem would be that new maps with new acc sliders would get a higher judgment for good accuracy, meaning that maps with simple rhythms will be even more overrated due to inflated pp from OD.
As the community is already heavily leaning towards pp-efficient maps this would probably discourage players to play "old" maps even more.

Endaris wrote:

smoogipooo wrote:

Please, do not discuss Star Rating and PP here.

It's been a year since v2 has been introduced and to me it looks like it has become a sideproject that is exclusively used for tournament scoring until osu!next is out. So please look at the topic in regards to how it affects tournaments and specifically OWC 2016, not how it affects pp/worse players/whatever.

Railey2 wrote:

Personally, I don't give a shit about my score, I just want acc and combo. I know many players think the same. I think if they handle it this way, there is not much of a point in implementing it. I think slideracc should affect the pp-system.

Railey2 wrote:

Why do you think that giving acc'd sliders as a tool to mappers would divide the community? If some mappers want to map "old style", they can do that by just mapping old sliders.

TakuMii wrote:

This entire thread is about improving the score system, not for complaining about the fact that it's not related to PP. Score V1 has nothing to do with PP for a very good reason, and that's because it's fundamentally flawed.

Seijiku wrote:

:?: So Do you prefer ScoreV1 or ScoreV2 ??

Vote here : https://goo.gl/rkPSDM

Result here : https://goo.gl/OQo1Wa

Seijiku wrote:

:?: So Do you hate change or do you blindly accept the future ??

JTF195 wrote:

I think slider ends should break combo like slider ticks do.
AFAIK, they're the only gameplay element that adds to combo but doesn't break combo. That isn't consistent.
It doesn't make sense that you can play through a song without a combo break OR full combo

JTF195 wrote:

I think slider ends should break combo like slider ticks do.

AFAIK, they're the only gameplay element that adds to combo but doesn't break combo. That isn't consistent.

It doesn't make sense that you can play through a song without a combo break OR full combo

JTF195 wrote:

I think slider ends should break combo like slider ticks do.

AFAIK, they're the only gameplay element that adds to combo but doesn't break combo. That isn't consistent.

It doesn't make sense that you can play through a song without a combo break OR full combo

Franc[e]sco wrote:

it will also prevent the development of bad slider habits in new players.

quaternary wrote:

But I can't think of a better way to score spinners other than making them not count at all, which is dumb. Maybe the difficulty to clear a spinner could be MASSIVELY increased so that it's actually really hard to spin a 300, and then remove spinnerbonus? Although that's an even worse solution. Idk.

den0saur wrote:

What will happen to fallback? Is it going to be implemented to fallback at least at any point in the future?

den0saur wrote:

What will happen to fallback? Is it going to be implemented to fallback at least at any point in the future?

Full Tablet wrote:

The score is supposed to indicate how well you played the map, not indicate how good the play was (that is pp's function).

In A FOOL MOON NIGHT, the play gets all 300s (awarding the base 1 million points + mod bonuses); while in Syrup, it also gets all 300s (awarding the million + mod bonuses), plus score from overspinning.

gellandor wrote:

Full Tablet wrote:

The score is supposed to indicate how well you played the map, not indicate how good the play was (that is pp's function).

In A FOOL MOON NIGHT, the play gets all 300s (awarding the base 1 million points + mod bonuses); while in Syrup, it also gets all 300s (awarding the million + mod bonuses), plus score from overspinning.

The point with puting those scores was to show how broke the system was. A map with such a low combo and skill requirement should not have a higher score amount (even just getting a 300 on the spinner would give you a higher score).

There should be some kind of score balance for higher maps so you feel like higher maps reward you better then lower maps (aka the higher star maps should NOT be 1m when you can get 1m on a 1 star)

TakuMii wrote:

I wouldn't worry about the ranking side of things; anything that'd be unrankable with ScoreV2 would also be unrankable under the current system anyways.

I mean, sure, it may make slider-intensive maps more frustrating, but it won't change anything gameplay-wise besides handing you more 100s than before (which isn't too big of a deal IMO, considering how combo still accounts for most of your score). SSes would still be possible on pretty much every ranked map (with the exception of a few incredibly old maps) even after slider-accuracy is added to the game. If anything, this should just cause the PP system to weigh wubwub maps even more than before, which would honestly be a good thing for people that are good at those types of maps.

-GN wrote:

Loctav wrote:

Fair question, I would have problems figuring out a proper multiplier for it (given we use scorev2). Would it be a positive one? (it obviously makes things harder) or a negative one? (since it makes you less likely to fail.. ok weak argument)

I am not an EZ player or have much insight in this aspect, so if you have suggestions in how to design it, feel free to propose something!

(without a multiplier, it would sort of lose its tactical purpose, wouldn't it? dunno...)

(i was gonna post this a week ago but forgot.)

It's a good question, actually. I would ideally prefer some kind of dynamic score multiplier(and i really think it should be made for every mod if scoreV2 becomes the official system), judging how difficult a map is to read/understand, but that's a bit too ambitious.

However, you should know that while EZ makes maps very hard to understand for the uninitiated, it also reduces the physical(aim/speed-wise) difficulty by a lot, and those that practice it can in most cases do better on maps with EZ on than they could without, given that they've practiced them. So having it give a positive score multiplier(or no penalty at all) would give those people a great advantage over others who don't play it on the regular, and i don't think that's the kind of skill that should be rewarded in a tournament like this.

I had trouble coming up with a reasonable multiplier for it myself, though. Using the old 0.5x multiplier wouldn't make it very useful for scoreV2 tournaments, i think. So i propose something like this:

Nomod gives 700 000 for combo + 300 000 for accuracy for a max score of 1 000 000.
EZ would give 550 000 for combo + 150 000 for accuracy for a max score of 700 000.

I estimate a 95% FC to be ~850k for nomod and ~625k for EZ with the current scoreV2 accuracy dropoff rate.

Obviously the accuracy part is lowered because of the low OD in comparison, but only reduced it to half the value, as to not devalue scores with high accuracies. Accuracy on EZ is more about the big mistakes you do rather than how well you follow the music, but it can still be hard to get above 99% on very demanding maps - and i think that's where picking EZ would be most helpful. In addition, picking EZ makes it easier to spam through harder parts of a map at the expense of accuracy, so i wanted to still keep it a little rewarding.

For the combo part - EZ consistency is really, REALLY hard to attain. Using it seriously in a tournament can be very risky, so i didn't drop it that much. For a tournament based score system, i think it makes sense. Compared to the accuracy part, i think most of the score should come from your combo when using EZ.

...so that's my proposal. Everyone loves EZ plays, so giving it a chance to be used on freemod picks might be good for entertainment's sake, at least, and with this, it'd retain some tactical value as well. It might be a little late to try to implement right before OWC, and it's not that important for most players, but if it's going to be a possibility, i'd want to see it executed something like this.

shortpotato wrote:

It feels to me (correct me if im wrong?) that the staff feel as if removing score V2 (or aspects of it) is a direct attack against all of the work they have done into making it.

chainpullz wrote:

Making slider acc into a difficulty increasing mod

Kyreo wrote:

Hey there.

I was somewhat disappointed by the new scoring system so I started to look for a topic which deals with it. I finally found it – that's why I'm here – so I have to share my point of view.

First, this system is somewhat discouraging to me. If you rock, it's ok. But if you fail...
I saw there's a sort of one million points limit. It is a huge problem since people might be discouraged to play harder maps. That's what I usually do to improve at osu! Recently, I got something like 90k points on a map rated "insane". 90k under 1000k... Though, my accuracy was about 82% or something. The score means I only cleared 9% of the map. Come on, that's unfair. How could you get some motivation to play & improve if the game tells you "yea, you've got 82% acc but you cleared 9% of the map, you fool". Am I supposed to play easy & normal maps until I get one million points on them...?

Second, the loss of points when you fail a circile, slider, of whatever.
It's about the same issue: progress. When you're on this slippery slope, playing again & again in order to improve, you sometimes get inspired by a divine might or something. It helps you to do a 200 streak combo or else, it shows how much you've improved. Then, a tricky pattern comes and you make some fails. Taadaa! You lost so many points, your last good move has been erased because of this loss combined to the fact there's a 1 million limit. Don't you think it's way too hard to climb back after a few fails? If the map is shorter than 2 minutes, you're screwed at the moment you fail a bit. Sure, it's a good thing for tryharding pros. But think about the little players, those who want to progress. It's pretty unwholesome...

Full Tablet wrote:

Kyreo wrote:

[Cut for obvious reasons; readability.]

The reason a 82%acc score only gives about 9% of the possible score is not because of the score cap, the same thing happens with scorev1 (a SS in a certain map can give you ~27 million, while a 82%acc score with misses can give you ~3 million, for example). In my opinion, this is not a bad thing. Getting through a map with ~80% accuracy usually means you barely passed it, so the amount of score you get from that pass should be close to the minimum possible; that way, scores are distributed more evenly over the possible range of values (instead of most plays be concentrated on the 90%-to-100% range, which would happen if the percentage of score was close to the acc% of the play)

In scorev1, the maximum score a map gives doesn't depend heavily on their difficulty (an Insane map only gives about twice the amount of score for the same amount of notes compared to an Easy map, based on the OD/HP/CS settings), it depends mainly on the amount of objects (the maximum score of a map is approximately proportional to the square of the maximum combo). In the same mapset, harder difficulties tend to give more score mainly because they have more notes, not because they are harder.

When you do poorly in a note (missing, or a bad judgment), the "current" amount of score decreases. In strict rigor, since the accuracy part of the score is not something that is accumulated (acc% is the average of the judgment values of the notes you have played so far), your score does not exist until you finish the map; the number in the corner can be seen just as an indicator of how well you are doing.

The "current" amount of score during play could be modified to be always increasing, by showing the amount of score you would get on the map if you missed every single note you haven't played yet, but that would case some issues (the acc portion of the score would be near 0 for most of the play, then jump quickly to the final value when close to the end of the map; during a multiplayer match or when comparing to previous plays of the same map, the current score value would be a worse predictor of how much score you will get on the map)

Philosofikal wrote:

You seem to misunderstand the purpose of the score system.

The score system is there to rank people as accurately as possible relative to each other. It is being changed because the current system provides very inconsistent information from map to map and rewards combo disproportionately over accuracy, not to mention the score difference between mediocre and amazing plays is marginal at best.

The score system is not there to make you feel good or encourage you (especially when you didn't actually achieve anything). That burden of motivation is on you and you alone. What you are proposing is essentially a watering down of the system so that everyone gets a participation trophy at the expense of compressing the scores into meaninglessness. You want to take away the impact of working hard for and eventually earning a great score so you don't have to feel bad about not being good at the game after eight whole hours of gameplay.

When everyone gets a reward, the reward is meaningless. Earn your rewards.

Kyreo wrote:

I'm not saying we should reward anyone, just that people need to be encouraged to play. The difference between what you call a mediocre and an amazing play will still exist whatever it takes. A very good player will do something like 980 000 to 1 000 000 points, that is to say an average achievement of 98% to 100% of the map while a "mediocre" (you're somewhat condescending just by saying so) with 500 000 points will have an average achievement of 50% of the map. It is totally legit to me. Besides, I'm not for "earning a great score", just something coherent with the true achievement. Having a 90k under 1000k is not representative of what I've done, clearly. It's representative of what I could have done a few days before when I was less experienced.

And yea, the scoring system is here to rank people. Still, it is more than possible while being logic at the same time. Scorev2 will not help e sport & competition at all. Why? Imagine a second: there's a 1 million limit. If someone already did the maximum score, then someone else, & again... How are these people ranked? Who will be displayed as the first? With the "limitless" system of scorev1, you could manage to surpass the first of the ladder by adding mods, doing better at accuracy or combos or by spinning faster. The limitless system is the one rewarding the best players who always manage to be in the ladder. With a one million limit, how many maps do you think we will have with numerous rank 1? Scorev1 already ranks the best players on top of the ladder.


Ps: I've a question; what about the leveling system? Since you progress thanks to the scores you make, will it be changed aswell?

Kyreo wrote:

I'm not saying we should reward anyone, just that people need to be encouraged to play. The difference between what you call a mediocre and an amazing play will still exist whatever it takes. A very good player will do something like 980 000 to 1 000 000 points, that is to say an average achievement of 98% to 100% of the map while a "mediocre" (you're somewhat condescending just by saying so) with 500 000 points will have an average achievement of 50% of the map.

Rayne wrote:

If you want to put on emphasis on accuracy which is undervalued in the present score system, you could just make the 300s be worth like 1k-1,5k points and make 100 / 50s still be the same. That way you can still surpass someone with a way higher combo than you by being way more accurate, while players that can hold both accuracy and combo will be rewarded with the maximum amount of points.

smoogipooo wrote:

Hey all,

We're just over two months away from OWC and I want to know what the community likes/dislikes about the current ScoreV2 system so we can get it perfected before OWC comes a-knocking.

THIS IS NOT FINAL
Please, do not discuss Star Rating and PP here.

Here are a few facts/talking points about the current osu! ScoreV2 system:

1. 70% of score comes from combo, the remaining 30% comes from accuracy.
2. Spinners award 500 points per tick.
3. The mod multipliers are as follows: HR - 1.10x, DT - 1.20x, FL - 1.12x, HD - 1.06x

Basically, I'm willing to experiment trying different stuff so throw out any ideas you have. I'll be checking this thread periodically but don't expect me to reply to every comment, and please _please_ don't spam me with PMs telling me to read a comment in here ;___;.

BWSnoob wrote:

Would love the option of score v2 on multiplayer mode, even if it's unranked

1. combo / (combo + a); where a is from open interval (0, ∞)
2. combo^a; where a is from open interval (0, 1) [basically a root function]

Eraser wrote:

Score wipe is still unavoidable though, if they want to fix the scoring system, or until they can find a way to convert all of the scores.

Z4ckFairX wrote:

In my opinion the new score system is just a percentage with way to many 0s after it.

Z4ckFairX wrote:

About the changes in slider accuracy I think that it's both good and bad. It's ok to demand of a rhythm player to play at the right rhythm at all times but what I actually don't get is what you gain from it. Every time i see a 100 on the screen after a slider i have to think what the hell i did wrong; did I click too early/late or not followed it correctly, also if you have no combo break sound or you just started a combo in the case of a 50 hit makes you wonder if you actually dropped combo or just mistimed it badly.

Philosofikal wrote:

Converting scores would be trivial. You just run all of the plays through the same algorithm used to calculate the play normally. There is no need to wipe anything, it's a straightforward, one and done conversion.

1. Only top 500 or so plays in the map leaderboard have their replays saved and you can't recalculate the score when you don't know where the misses/50/100/300 happened.
2. AFAIK osu! times your hits down to the millisecond, but replays are recorded only at 30fps (60 for newer ones). So even when you have a replay it still isn't enough to tell when hits happened.

Only top 500 or so plays in the map leaderboard have their replays saved and you can't recalculate the score when you don't know where the misses/50/100/300 happened.