It's been a long while since the !mania mode has been introduced to osu and a quick look at the rankings for maps, whether they are easy maps, average-player level maps, or hardest-maps-available, in literally all of those cases one thing is clear:

Nobody uses mods for ranking.

NOBODY.

It's not even scarce, it literally doesn't exist. The only mod plays you'll ever see anywhere are a random DT score once in every 5 hardest-map scoreboards by players who have beatmania experience and did those scores effortlessly for fun, most likely in a random multiplayer match.

As it stands right now, it's possible to full-score (1000000) a map nomod.
The logic behind mods is that "it's still hard to get this full score nomod, and mods will make it slightly easier". What the hell, seriously. This line of logic ignores a very basic principle which is the fact that mods like HR and DT make maps about 100 times harder and, that players who are able to play maps with those mods and get anything that resembles a decent accuracy with those mods which would be good enough to rank with can pretty much full-rainbow-SS those same maps nomod effortlessly.

Maybe for HD and FL the current mod logic can somewhat hold since they're different types of difficulty, but for HR and DT, mods are practically worthless for ranking. There is no reason anybody would ever use HR or DT to rank on a map in mania, ever. Those mods right now only exist as a "for fun" option, which is not only a waste of a feature, but it also makes scoreboards in mania completely stale.

Quick recap: The problem is - mods (most specifically, HR and DT, but the same is pretty much true for HD and FL) are useless for ranking in osu!mania.
Scoreboards are stale.

Proposed solution: Give players an incentive to rank with mods. The only reason anybody would want to rank with mods, that make the map much harder to play (which means that, if you're able to play that mod even half-well, you can usually play the same map perfectly in nomod) is if mods give you a bonus over nomod plays. So basically, give the mod multipliers back. i.e. +6%, +12%, for potential higher than 1000000 scores.

What woc is going to say: "but the idea is that you can tell from the score how far you are from perfection on the map, and that 1000000 = perfect". This is nice, BUT, a map on DT is a completely different map, this alone means that DT has to give something -higher- than 1000000 no matter how you look at it. As for HR/HD/FL, the way that HR increases the OD means that the difficulty of getting accuracy skyrockets compared to nomod. With the nature of how players get accuracy (ms deviation from hits), somebody who can get decent accuracy with HR on a map will be able to SS that map completely effortlessly nomod, with no exceptions. Thus, HR has to give a higher total potential score than nomod or else it's completely worthless to use it.

As for HD and FL, I don't even get what's the point of HD right now considering you can control your scroll rate and hit offset which literally means that the HD version that hides the top is pretty much nothing (it's on the same level as changing the scroll rate and playing on a bigger window, lol) and the version of HD that hides the bottom is pretty much just "click things with a delay", which only ever matters on maps with SV changes and if there are no SV changes this is as good as changing nothing in the map. And FL is funny because it's basically just both of the HDs combined.

Would like to hear what better more experienced mania players have to say on the matter, and hopefully this can lead to some change, because undeniably nobody uses mods for ranking in mania ever since it's been introduced and it's straightforward mathematically pointless to do so.

no mod multipliers

1.mod still help for your pp increase.
2.I won't change mod multipliers because there's already millions of scores with or without mods.
3.When I see enough full score, i'll implement one more mod with non-upper bound multiplier, but not for now.

woc2006 wrote:

2.I won't change mod multipliers because there's already millions of scores with or without mods.

Most (most likely all) mod scores are either random plays or plays for fun without the intention to rank, I doubt any mania player would feel betrayed if suddenly new mod scores started giving more points. If anything, it would just give people an incentive to play more on the ranking.

Also, you won't see "many full score" any time soon, for the following reasons:
1. PP-gain is related to rank on a map. This means that for somebody who is playing for rank, it's more efficient to play new maps that barely anybody played instead of trying to improve their play on a contested map. Since there are 7000(?) maps in osu now, you won't see people caring to full score maps in at least a year.

2. Players who are actually good enough to full score a map don't have fun doing so, since the map is way too easy for their play level, and they'll end up just playing harder things and eventually going to unranked maps since ranked maps pose no challenge, nor does ranking on them with the current system due to lack of a point to mod usage.

Why are you so strongly intent on not giving mods bonus multipliers? Every single mania player I've talked to with no exception says they hate the current way mod bonuses are implemented and would prefer to have a score bonus.

Right now, mods only reduce "punish", which only occurs if you get 200 or lower - players with SS would get the same score with or without mods. Why not make the mods give you slightly more score on 300 as well, so scoreboards could be more balanced something like so:

#1. MAX:940, 300:10
#2. MAX:900, 300:50, +DT
#3. MAX:920, 300:30

Millhiore the point is that as long as you can't get over 1000000 with mods, you have no reason to use mods ever, because getting 1000000 without mods will always be easier than getting 100000 with mods no matter what multipliers you give them. Or at least, this is undeniably true to HR and DT on a basic mathematical level (it's physically impossible for a player to be able to perform well on a mania map with HR or DT and not be able to SS it perfectly nomod). HD and FL are kinda iffy at the moment for this game mode anyway.

Indeed, having 1M as the max score is a bit weird, but max score is barely even achievable. Making 300s give more score with mods would be a compromise, since I don't think woc wants to increase the max score, and this would give SS players a reason to use mods (there's LITERALLY none right now if you SS, while there's at least a negligible score boost for the same accuracy with mods otherwise)

MillhioreF wrote:

Indeed, having 1M as the max score is a bit weird, but max score is barely even achievable.

Except,

- Getting full rainbow SS on a map that you can play well is actually NOT that hard, and anybody who'd care to retry to do that can do it.
- Getting a rainbow SS or even an SS at all with HR, takes the same skill-set as doing that on nomod, only in a 100x higher level, this means that somebody who can do it with HR (or even do like 95% with HR) can get that rainbow SS nomod EFFORTLESSLY because it requires the same skill set that he has on a much higher level.
- Do I even need to explain DT? :v

Aqo wrote:

MillhioreF wrote:

- Getting full rainbow SS on a map that you can play well is actually NOT that hard, and anybody who'd care to retry to do that can do it.

Yes it is. Unless you're talking easy mode, then who cares?

Using mods for score bonus is a stupid concept to begin with.

I've seen max score every now and then, but only on Easy diffs. Can you link me to a Normal or higher that someone maxed out the score on?

MillhioreF wrote:

I've seen max score every now and then, but only on Easy diffs. Can you link me to a Normal or higher that someone maxed out the score on?

obviously no one is going to bother doing it so there will be no max scores there, since if you're at the point you can max score those maps you'd prefer just playing harder maps

ask any decent mania player to try to get that max score on a random map and with retries he'd be able to do it. except why would he do it normally in his usual play routine? there's no point. rankings have no real competition right now, and the way that PP works even encourages having no competition on rankings - instead it encourages you to spam easy plays on many maps.

---

no matter what anybody says
osu!mania has been out for a while
nobody uses mods for ranking
nobody likes the way mod bonuses work right now
there is clearly an issue that needs to be addressed. you can't just shrug it off and say "oh it's fine due to some theoretical logic that doesn't work in practice"

Aqo wrote:

and the way that PP works even encourages having no competition on rankings - instead it encourages you to spam easy plays on many maps.

This is a problem with PP, not mania mods.

Aqo wrote:

ask any decent mania player to try to get that max score on a random map and with retries he'd be able to do it. except why would he do it normally in his usual play routine? there's no point. rankings have no real competition right now,

Yes they do, SS scores beat other SS scores based on number of rainbow 300s. You want 1st? Get more of them. Show me a single remotely difficult map with even a single 1,000,000 score. Keep in mind that virtually every auto-converted map is easier than max difficulty charts made specifically for mania.

While you're at it, show me an osu!standard map with a massive 100% SS tie at OD10+. If you can do that, we can then claim that osu!standard is noncompetitive because you can't beat those scores. Insane difficulties have an even tighter timing window for rainbow 300 by default, you don't need silly mods to do it for you. Oh, and "slider" timing actually matters here too, so it's even harder.

Aqo wrote:

nobody uses mods for ranking

That's because all the mods suck for this mode. Games like IIDX and pop'n music and whatever have some cool mods, this game has nothing interesting (for this game mode). Even in those games, people only really use them to make things easier on them (speed changes, mirrored/random note placement, cutting off the top half of the screen or sudden, etc).

So Lyby, do you feel that the current way mods are implemented in mania are fine and better than a score-increasing system?

This is less of an issue now, but when mania-specific maps start getting ranked it will become a major concern.

If you feel that the current system is good and that giving score bonus would worsen it, please say so.
Everything you're saying so far is just trying to explain the theoretical logic that led to the current state of the system.

Simply answer these:
1. Would you start using mods more if they gave a score bonus above 1000000?
2. With the current system, do you see yourself choosing to compete for a top score on a map with HR or with DT, instead of by retrying nomod, sometime in the future (ever)?

Forgive me but I fail to understand why are you trying to defend a system that has proven itself as a failure in practice when there's another much simpler system that has already proven to work.

the best implementation of mods would be zero effect on score.

This won't be an issue on mania specific because the maps will be so much more difficult that no one will ever get 1,000,000 on insane difficulties.

I don't understand why you keep bringing up HR. That mod is utterly pointless in this mode, because the timing window is already tight by default, and it doesn't even change anything else.

Lybydose wrote:

I don't understand why you keep bringing up HR. That mod is utterly pointless in this mode, because the timing window is already tight by default, and it doesn't even change anything else.

To give an extra reward for the players who are able to perform well on maps while still hitting the incredibly tight window and YES players like that exist or will emerge eventually.

maps that were created for o2jam as impossible joke maps were still later on AAA'd by players, everything that looks impossibly hard now will seem trivial later as competition grows. why do you want to restrict this

Lybydose wrote:

the best implementation of mods would be zero effect on score.

I don't understand this statement. This means you're personally against difficulty-altering mods and only in favor of "style" mods. However a lot of players enjoy difficulty altering mods for the extra difficulty which, in a way, creates new maps for you to play on once you get to a high enough level. But with no reward this remains as a "for fun" option, while for many players ranking is fun; why not allow this to be useable for ranking? Even the players who can't rank with them find it interesting to see the competition between those who can.

Right now on a diff spread of HD/IN/EX, IN+DT is often harder than EX, and it's like the "next difficulty level" of the mapset. Why can't this be represented in ranking too to motivate players to try to perform well and compete for that as well?

Ok, let me explain why HR is utterly pointless here:

It doesn't make the map any more difficult. At all. All it ends up being is a binary "click this if you want bonus points" button. It doesn't matter that my accuracy goes down, I get a bunch of bonus points because I hit a magic button before I played the map. EDIT: My knowledge of Taiko is limited, but it seems scores are already pretty much like this. You hit Hard Rock on everything you play for bonus 6% points because it doesn't do anything to make the map harder, and the tighter accuracy doesn't offset the 6% gain.

There's already an "extra reward for hitting incredibly tight timing windows", that's the entire point of rainbow 300s. If you think those are too easy, see about getting THAT changed, not implementing a hard rock option.

I don't know or care what AAA in o2jam means or how tight the timing is in that game. I do know that in the hardest IIDX maps, no one ever gets 100% perfect and it's plenty competitive without adding in ridiculous mods. They do get AAA, but "AAA" doesn't mean "perfect every single note".

Play with mod > Play without mod with same score. It would be nice if tied scores were given to the modded plays. If you want to top that 1000000 for number 1 spot, play with a mod to beat it. That would make mods more useful IMO.

So here's a new suggestion Lyby, since I do kinda agree with your view on HR
get rid of all mods for mania
add only the following mods:
1. key change mod (4k~8k same as current one)
2. speed mod, letting you use any multiplier from 0.5x to 2.0x with 0.1x increments
every 0.1x increment upwards from 1.0x gives +0.5x score (so that current-DT 1.5x is +25%, and 2x speedup would be +50%), and every one downwards is locked to exactly 50% score regardless of the increment

how's this?

Even while I would like that speed-up for standard (since scores are highly contested) I'm not sure about how big osu!mania is and if that would really make any sense. Like I can see people farming #1s by just adding the minimum multiplier.

Not a mania player tho so feel free to ignore me.

Wishy wrote:

Like I can see people farming #1s by just adding the minimum multiplier.

???
mania revolves around accuracy, not FCing. If your score in halved to 50% you can't really farm anything with this.

What I mean is if x1.1 speed gives some considerable bonus, I can see people adding the minimum multiplier to get a high score. Just like you beat some HR high acc on standard by just using Hidden and getting a few less 100.

Wishy wrote:

What I mean is if x1.1 speed gives some considerable bonus, I can see people adding the minimum multiplier to get a high score. Just like you beat some HR high acc on standard by just using Hidden and getting a few less 100.

Well yeah that's the idea. Everybody would play with the highest multiplier that they can play with and still get decent accuracy on, and then scoreboards will actually be competitive in the sense of who can play the map layout and patterns the best.

They would play with the minimum multiplier required to get #1, until some pro comes and uses a very high multiplier, which according to what you told me is not likely to happen since the best players don't really rank stuff.

It's been fairly common in stepmania to play songs with speed multipliers like that because if you can AAA something at 1.1x, and someone else can AAA it at 1.5x it's easy to tell who the better player is, even if both of them could also AAA it at 1.0x. It allows for a finer grain way of telling who is actually better at that map, because the faster the song, the harder it gets in a pretty straightforward fashion.

If we're discussing mods and suggestions for the mod system in general, I'd say that we should also have some sort of pattern randomization. Its been in basically every mania game because it gives a HUGE boost of replayability on maps. I think that mods like that however, should have a separate leaderboard, because sometimes a map can be easier when generating randomized note positions... But I still think that it should be available (and should have a leaderboard all it's own, so you're not competing with the people playing the standard chart). The bonus of it having a separate chart is that there's no pressure from trying to compete with scores on the standard map which may not be an incentive, but at least it's not got the negative aspect of trying to compete with standard scores.

You know, those maps you call "joke maps people later on AAA"

did you know there are human limits?

Human limits are always higher than people expect though... I mean, if I hadn't seen it myself, I would have said nobody can be as good as Entozer, and yet there are people even better than him...

Osu!mania with mods can be more interesting for example look how Remilia Scarlet can play with DT or something like that 8D

Bobbias wrote:

It's been fairly common in stepmania to play songs with speed multipliers like that because if you can AAA something at 1.1x, and someone else can AAA it at 1.5x it's easy to tell who the better player is, even if both of them could also AAA it at 1.0x. It allows for a finer grain way of telling who is actually better at that map, because the faster the song, the harder it gets in a pretty straightforward fashion.

The speed Remilia-Scarlet chooses to play is nearly 0.7x of me. He/She can SS a map and I may also be able to do so. Who is stronger?
You can't decide a player's strength by the speed he/she choose. Actually one 1.0X SS is far easier than an 0.05X SS.
Or you may be talking about bpm-raising on actual song. Though DTing songs make it really hard, you still should not make decisions on multipliers here. The reason is simple: If I can endure 300bpm 1/4s, I'll just raise every song's tempo to 300 to maximize my point profit - That's ridiculous.



And there are a lot of reasons that we don't use score multipliers in o!m, in our early discussions with woc.
1.If you have HR*1.06, everyone will start playing HR, and finally things become no mod *0.94+ (And become "no mod - a little easier judgment, a little more tolerance").
2.Granting these mod multipliers would be just result in a gift for pros and a punishment for beginners.
3.A few mods combined would just be unplayable because players can't see anything on the screen. Granting score mod for these mods just make cheaters want to cheat. (like HD+FL in taiko)
4.In standard, pro players always start playing a map with HD, using transparent skin for that they can 1 pc play beating no-mod. You may know that it's why osz2 come into place.
5.There's already examples in early games that use mod bonus like DJMAX Online. In DJMAX everyone played with S.Random(A mod that shuffles the whole map) for 10k bonus (That game had a score cap of approximately 240k, with higher judgment bonus score - but with 100% you still have a cap.), nearly 1.05x in hottest competition. Since then no one played original map.
6.The last reason comes from woc's designing conception. Higher judgment = Perfect. Eliminating ten 300 is FAR MORE IMPORTANT THAN any kind of crazy play. (That's why MAX(300g) judgment is set to extremely hard level.) You are seeing a few 1ms on easy maps, but it's not so in mania-specific ones. Like others said, KEEP IN MIND THAT mania-specific maps would be extremely hard. With current score system, one 1% judgment plus would be nearly 1.05x at S, so we don't think that it would be not competitive on these maps. (The score system had been modified once in close beta test just to ensure this.)

And you may say that "mod makes game harder why there wouldn't be a multiplier?"
For this point, standard game play is standard play. it's really hard to judge whether a multiplier should be applied for a mod due to its "difficulty". I already see a couple of players: They play FL at their own favor, they play poorer in no-mod. Same thing come into DT: A few map is not becoming a lot harder (Sometimes easier due to speed limit!) in DT. If you grant score for these mods you are just forcing everyone to play samely, with unbalanced score multipliers. No-mod would become a historical thing since no one plays it.

So after a long discussion with woc we decided to deny any mod multipliers for "diff-increase" mods before public scoreboard is released.
Enjoying natural game play is good, why not play no-mod?

1- HR requieres your accuracy (timing) to be better since it increases OD. HR should not have a big bonus as it has on standard yet it makes the map harder to get high acc on so it should give a bonus. It's harder to SS OD 10 than OD 8.

2- What's wrong with that? If I'm good and can DT everything then I should get the top scores. If my accuracy is so perfect I can SS even on HR, then same thing, I should be on the top. It's not about punishing beginners but rewarding pros. Your idea could be used pretty much for hard maps so please don't rank ANY HARD MAP I CAN'T DO because you're punishing me, a beginner, who wants to rank over the pros even tho I can't play this mod at all.

3- Don't add mods like FL and HD. FL = cheat mod and HD is kind of dumb for mania I think.

4- Wrong, you should play standard before talking. HD makes stuff harder not only because of stacks (that you might read first play without transparent skin) but because of many other things you gotta play to understand.

5- Because the map was too easy. If you're good enough to constantly SS something, why not be able to use mods to get a higher score? If you want to avoid that just get hard maps.

6- Then allow the usage of HR or custom OD. If I can SS OD 10 I should get a higher score than SS OD 8 because what I'm doing is harder. Same thing if I can DT some map and get SS and you don't then my score should be higher. What you can do is give mods a not very high multiplier so you still need to be very accurate for them to give any bonus.

This is a bad example but:

http://osu.ppy.sh/b/198271

If you use DT, it is harder to get accuracy than nomod, then why is DT SS worth the same than nomod SS?

Wishy wrote:

If you use DT, it is harder to get accuracy than nomod, then why is DT SS worth the same than nomod SS?

DT doesn't actually change OD at all in mania. The notes just come faster and closer together.

Wishy wrote:

1- HR requieres your accuracy (timing) to be better since it increases OD. HR should not have a big bonus as it has on standard yet it makes the map harder to get high acc on so it should give a bonus. It's harder to SS OD 10 than OD 8.

2- What's wrong with that? If I'm good and can DT everything then I should get the top scores. If my accuracy is so perfect I can SS even on HR, then same thing, I should be on the top. It's not about punishing beginners but rewarding pros. Your idea could be used pretty much for hard maps so please don't rank ANY HARD MAP I CAN'T DO because you're punishing me, a beginner, who wants to rank over the pros even tho I can't play this mod at all.

3- Don't add mods like FL and HD. FL = cheat mod and HD is kind of dumb for mania I think.

4- Wrong, you should play standard before talking. HD makes stuff harder not only because of stacks (that you might read first play without transparent skin) but because of many other things you gotta play to understand.

5- Because the map was too easy. If you're good enough to constantly SS something, why not be able to use mods to get a higher score? If you want to avoid that just get hard maps.

6- Then allow the usage of HR or custom OD. If I can SS OD 10 I should get a higher score than SS OD 8 because what I'm doing is harder. Same thing if I can DT some map and get SS and you don't then my score should be higher. What you can do is give mods a not very high multiplier so you still need to be very accurate for them to give any bonus.

This is a bad example but:

http://osu.ppy.sh/b/198271

If you use DT, it is harder to get accuracy than nomod, then why is DT SS worth the same than nomod SS?

Actually In the early stage of discussion I stand on the side of HR adding points like you for the same reason as you, but after a quick thought I immediately give up. It's easy thought - using HR adding point = at the same time SS, get less 300. So there's no need adding points on HR.

First target = getting S (800k)
Second target = getting SS (995k) --> normal play
Thrid target = getting low 300g (998k) -->hr play

Actually the last 3k matters! Just like mods in other game mode: If you want to get #1, get 1 more 300g. (That's why Remilia-Scarlet is on the top - he/she plays extremely accurately, getting a great amount of 999k/998k #1s)

You can DT, you can SS easy maps, but your judgment may still be poor. If you get an SS or whatsoever (on HR or DT or anything) and get less than 998k, don't think that your accuracy is high! The spirit of o!m scoring is accuracy, not crazy playing.

For HD in standard I was just trying to represent a possible situation of conflict between no-mod players and mod players on game preference. No one want to see that kind of problem happing again, right?

Thing is if you play HR, meaning higher OD, and you get 97%, maybe that 97% HR play may be SS nomod (because of timing windows 100 on HR may be 300 nomod). Maybe just giving a very small bonus to HR plays (1 or 2%?) could work. Like you still have to get very high accuracy (99%+) to bear SS nomod.

Maybe letting players just change OD and get bonus score if they increase it could work too. Getting little bonus for higher OD because accuracy is harder. If you get SS OD 10 you should not get the same score as SS OD 8 . Would be a good challenge for pro players.

Wishy wrote:

Thing is if you play HR, meaning higher OD, and you get 97%, maybe that 97% HR play may be SS nomod (because of timing windows 100 on HR may be 300 nomod). Maybe just giving a very small bonus to HR plays (1 or 2%?) could work. Like you still have to get very high accuracy (99%+) to bear SS nomod.

Maybe letting players just change OD and get bonus score if they increase it could work too. Getting little bonus for higher OD because accuracy is harder. If you get SS OD 10 you should not get the same score as SS OD 8 . Would be a good challenge for pro players.

In mania it's not like in standard, for there's an MAX(300g) judgment. It's far harder than 300. say, 300 is od8(a little easier) and 300g s od10(actually a lot harder). So there are already no need for an HR mod - you play no mod and at same time score as HR (Actually MAX as 300 is x1.06!)

So if taking standard as an example, SS an od10 grants far higher points due to 300g. if you can only SS od8, you may suffer from nearly 3% point loss.

Oh I see, didn't know that 300g was so hard.

xxbidiao wrote:

In mania it's not like in standard, for there's an MAX(300g) judgment. It's far harder than 300. say, 300 is od8(a little easier) and 300g s od10(actually a lot harder).

Are you sure about this?
Would be nice if some dev posted the actual numbers. With Unstable Rate 150, about 66% of the hits will be rainbow 300s and the rest regular 300s, while on standard on OD10 you'd get about 87% accuracy on circles-only with the same unstable rate.

Hm I believe that unstable rate is the sum of the difference in ms you had on hits that deviate from the timing @0ms. Don't think unstable rate matters that much when it comes to extreme OD, since it's a sum instead of a linear statistic. Don't quote me on this though, I could be wrong.

But yes I can vouch for 300g to be of OD10~ or maybe even harder difficulty to hit, while regular 300 feel like something between OD7 and OD8.

lolcubes wrote:

Hm I believe that unstable rate is the sum of the difference in ms you had on hits that deviate from the timing @0ms. Don't think unstable rate matters that much when it comes to extreme OD, since it's a sum instead of a linear statistic. Don't quote me on this though, I could be wrong.

On two maps I had the following :
99.5% @ OD7 with 148 unstable rate
99.18% @ OD9 with 178 unstable rate
So I guess it's something like that

oops, i quoted you

Aqo wrote:

xxbidiao wrote:

In mania it's not like in standard, for there's an MAX(300g) judgment. It's far harder than 300. say, 300 is od8(a little easier) and 300g s od10(actually a lot harder).

Are you sure about this?
Would be nice if some dev posted the actual numbers. With Unstable Rate 150, about 66% of the hits will be rainbow 300s and the rest regular 300s, while on standard on OD10 you'd get about 87% accuracy on circles-only with the same unstable rate.

Unstable rate is standard variance. So it counts for nothing. (You may say that normal distribution works, sometimes it do! but not if just a few notes slipped from 300/200. In your examle at unstable rate 150 if you are playing a pure circle map, you should have nearly 80% rainbow 300s on SS using ND as predictor, assuming that you have adjusted your offset and made hit error number added to 0. (-5ms~+5ms like that) Otherwise you may have far lower accuracy. Slider judges a bit differently, so you're harder to get 300g on that.)

And I'm pretty sure about my judgment comparison there, though I'm not sure about the exact number (Precision of 1ms).

300g:    0~15
300:     16~34
200:     35~66
100:     67~96
50:       97~120

Approximate MS values for different judgements on mania nomod. May be off by 1 or 2 ms (tested by Winshley, confirmed by me)
Rainbow 300 is slightly stricter than OD10 300, normal 300 is slightly stricter than OD7 300, hitting at all is slightly more lenient than OD8.

We're still working on figuring out the exact values for HR and Easy. DT and HT don't actually change the judgement windows, it just makes the notes closer or further from each other.

xxbidiao you quoted me on page 2 and I wanted to clarify what I meant. When I mentioned playing a song on 1.5x etc. I meant song speed, 1.5x speed would be DT in mania. I also want to point out that I did not specifically say I wanted multipliers on mods. I don't actually want mod multipliers because frankly I think its a bad idea that just encourages people to use mods on everything. However, you also said that always adjusting songs to be at whatever speed you were capable of is "ridiculous". I disagree. If I can stream faster than you, I'm better at fast stuff than you, so I should be ranked higher accordingly, unless you are equally better than me at something else. If a player can play a song with a mod that makes it objectively harder, then they deserve to be rewarded for that increase in difficulty (assuming a perfect ranking system).

However osu's ranking system is far from perfect:
Pros in osu play with mods pretty much always on. Why? Because they can, AND because there's no reason not to, and bonus score for leaving it on. In many other games, mods are rarely used except when the player wants to mess around, or if the player wants to show off in some way (beat the song at 1.5x, sightread it with random notes, etc.) It's easy to farm scores if you have lots of beatmaps, PP or not. I could probably be top 50 in mania if I sat down and ranked scores on every standard map I own. It would be even easier if I got a substantial score boost from playing with HR or some other mod, or a mod combination.

What nobody has mentioned is the impact that mod multipliers have on the beatmaps themselves. In standard, it is extremely rare for a map to be so difficult that only the absolute top tier players can pass it. If it is that hard, it's almost definitely not ranked. For top level pros, the only way that 95% of ranked maps are interesting to play is with mods, because otherwise they're just too damned easy.

If you make is so that mods are no longer necessary for people to rank highly, it means that you can't get away with making an "insane" that most decent players can FC easily, because frankly, if I can FC something, Entozer can get 99%+ (hell, even if I can't FC it, if i can pass it' there's a good chance he will get 99%+). With mods no longer contributing to your score, mappers will need to step up and actually make hard maps. Its by no means a guarantee, but it's more likely to succeed in mania since we already have mappers who are making difficult maps.

Something else I should point out is that standard maps in mania are easy, period. Mods on standard ranked maps is like playing HD+DT+HR on a Medium with large circles, low AR and low OD... The mods make it harder, but not enough to matter to anyone who's actually good at the game.

So, here's what I believe:
1) Mods make things harder, and people should generally be rewarded for doing good on harder stuff.
2) The current ranking system rewards people for placing moderately well in a large number of songs. The more, the better. (farming)
3) If mods give you more points, it makes farming easier (as long as the extra points outweigh the impact of the difficulty increase of the mod)
4) Most standard maps in mania are EASY. Mods that give points just encourage farming with mods because of how easy they are.
5) If mods don't increase your score, it will be up to mappers to make up for this gap in map difficulty.

Suggestions:
Mods should not contribute in any meaningful way to your score.
There should be separate rankings, or some way to ensure that people who farm standard maps in mania do NOT get ranked along side players who are good at mania.

MillhioreF wrote:

300g:    0~15
300:     16~34
200:     35~66
100:     67~96
50:       97~120

Approximate MS values for different judgements on mania nomod. May be off by 1 or 2 ms (tested by Winshley, confirmed by me)
Rainbow 300 is slightly stricter than OD10 300, normal 300 is slightly stricter than OD7 300, hitting at all is slightly more lenient than OD8.

We're still working on figuring out the exact values for HR and Easy. DT and HT don't actually change the judgement windows, it just makes the notes closer or further from each other.

My data is coming from woc. Don't tell him that I said these here
300g = 16/17ms, I forget it.
300 = 33/34ms
200 = 65~70.
other missed
And if you hit too early, miss = 120-150? I forget to remember lower-than-200 things...
Woc has mentioned that if you hit too late you won't get 100 and 50. I didn't test that.
Slider are judged as average hit error of start point and end point. and you are more forgiving in slider judgment. (How forgiving? woc didn't make it completely clear, but it is 1ms in 300g)

HR = /1.4 (300g = nearly 12), EZ = *1.4. down arrow makes 300=47,200=77 (No change on 300g).


Bobbias has mentioned me in the previous reply.
Sorry for my misunderstanding to your post, though I have already clarified that in my post before.

I held the same view as you that no bonus score should be awarded by mod playing. I said in my post that raising song BPM is "ridiculous", and actually it's not standing for "A higher BPM of the original song is not hard". Players would like to play another harder song (or just the higher difficulty of the song) to show their level. In this point, raising BPM crazily is meaningless and "ridiculous". Imagine if this dialogue would happen:
A:"I played osu!tutorial in 1000 BPM"
B:"Wow you are so strong!"
I believe that no one want to see this ever happening!

xxbidiao wrote:

I held the same view as you that no bonus score should be awarded by mod playing. I said in my post that raising song BPM is "ridiculous", and actually it's not standing for "A higher BPM of the original song is not hard". Players would like to play another harder song (or just the higher difficulty of the song) to show their level.

This depends on how mania maps will be ranked, though. At least in standard, right now, due to some silly reasons or whatever, truly challenging maps are close to unrankable. The standard equivalent to Entozer's [Insane] maps has zero chance of ever getting ranked, so players have no choice but to use mods to play the stuff that's fun for them. If mania would allow for maps like that to get ranked then mods might really be unnecessary as there will always be an ongoing nomod fight.

I tested with HR as well. The sample is not complete, but I got 0~11ms for 300g with that mod.

Yes, to get MAX / 300g, you must not time any earlier or later than 11ms from the note. Basically it's 1ms (or 2ms if counting both the early and late together) stricter than OD11.

Need to note that this may be inaccurate and/or subject to change though. It's better if we could just ask woc2006 anyway.

While I could understand that HR gives less score punishments because of this, I do find it funny when people somehow managed to score all MAX, which is 1,000,000 as well. Then again, getting all MAX simply is just too hard on harder maps anyway. Which is why I'm indifferent about the change.

Aqo wrote:

xxbidiao wrote:

I held the same view as you that no bonus score should be awarded by mod playing. I said in my post that raising song BPM is "ridiculous", and actually it's not standing for "A higher BPM of the original song is not hard". Players would like to play another harder song (or just the higher difficulty of the song) to show their level.

This depends on how mania maps will be ranked, though. At least in standard, right now, due to some silly reasons or whatever, truly challenging maps are close to unrankable. The standard equivalent to Entozer's [Insane] maps has zero chance of ever getting ranked, so players have no choice but to use mods to play the stuff that's fun for them. If mania would allow for maps like that to get ranked then mods might really be unnecessary as there will always be an ongoing nomod fight.

From what I've seen reaching the maximum score (meaning 1m) is kind of hard and it most maps I just find scores close to that, meaning you can still compete. Thing is I'm not sure about how hard they try to get those close to 1m scores, meaning I don't know if it's really that challenging for good players.

Wishy wrote:

From what I've seen reaching the maximum score (meaning 1m) is kind of hard and it most maps I just find scores close to that, meaning you can still compete. Thing is I'm not sure about how hard they try to get those close to 1m scores, meaning I don't know if it's really that challenging for good players.

if 300g is at least as hard as OD10 (if not even harder) i think it is

Instead of how hard it is to grind a map until you 1m it due to 300gs, you should ask yourself if this is more interesting for players than having a competition between 1.x DT multipliers and who can play the map the fastest while still performing well on it.

I agree with that. Retry spam on standard is boring as hell (and it's pretty much retrying until you don't get any random 100).

GoldenWolf wrote:

if 300g is at least as hard as OD10 (if not even harder) i think it is

Based from my no-mod sample (as posted by MillhioreF), the timing for getting MAX / 300g is same as OD10.5.

Wishy wrote:

I agree with that. Retry spam on standard is boring as hell (and it's pretty much retrying until you don't get any random 100).

It's still a very good practice. With time your senses get better as well.