[Proposal] osu!mania Ranking Criteria (whole draft)

posted
Total Posts
56
Topic Starter
Feerum
Hello everyone!

Old thread gone and here we have a new one. We re-worked the old draft based on the feedback from the old thread and are ready to publish it once more for discussion. You can find it here



As usual, please leave feedback in this thread. Everything helps! Speak now or forever hold your peace.

The thread will be open for discussion until the 17th December, 23.59 UTC+0.
After that we will close it and re-work it once more if needed or finalize it as our new osu!mania Ranking Criteria.

Happy discussion!
pishifat
this time the draft won't fall inactive because i purchased some shiny new whips
Hydria
rules guidelines are made to be broken
Xonica
"Long notes should not have release timings that use different timing snaps."

What is the purpose of this? If it's technically accurate, and the note in the song correlating to the long note has the same begin and end points, this will not improve the overall quality of the beatmap. If anything, I feel that this guideline will make maps of certain songs unnecessarily hard to get ranked.

"Avoid 1/8 streams lasting longer than 4 beats."

4 beats in particular seems like an entirely arbitrary figure, which again will limit the creativity of mappers, and this can be said for all the other arbitrary guidelines in this proposal. I agree with limiting the difficulty of an insane map, but the difficulty of a file is not only dependent on the length of the stream, but also the bpm and patterning.

The one good thing going for this proposal is the lack of limitations on expert difficulties. I think I speak for everyone when I say that the current formula for ranked maps is incredibly stale, and anything which allows for more freedom and creativity from mappers is a good thing.
Tailsdk
I feel like the long term SV for expert difficulties could be a bit bigger imo, but i guess its just a guideline so i dont mind it.
And maybe something on the criteria that tells peple about how normalized BPM SV values effects the recomended long term SV values. Since that is technically a long term SV.
Everything else seams fine to me.
EDIT: i agree with Xonica on the LN snap
UndeadCapulet
std person here, just popping in to request y'all try to stick to glossary terms being singular instead of plural, for consistency w/ the other sections of the rc. ty~
MEGAtive
I think Long Notes may be released at different time if the deviation between each releases is pretty long. I remember some music have two distinguishable string instruments, one ended a measure or two later than the other. That, I think is still manageable for newbies. Probably add some exception for LNs with releases deviation exceeding 2 or 3 seconds for Easy and 1 or 2 seconds for Normal. Or just use how many measures or beats are acceptable for LNs to have different release time.
An exception to this rule would be the usage of grace notes that are at higher-end snaps (1/8, 1/12 and 1/16) as they are only used to accentuate gracing sounds.
I think stating how much note a grace are acceptable at this difficulty would be beneficial. Or how it correlate with the lesser difficulty. Hard with 1/8 grace with three notes would felt like a spike if the Normal counterpart only uses single notes or doubles.
Arzenvald
[img:1337][/img:1337]Can we finally have 7 simultaneously notes rankable?


5 consecutive 1/4 notes
Is this mean we can put 5 1/4 on faster bpm song? which certainly not cool, Or is there any certain bpm threshold for this?
Kawawa
Difficulty Specific
- Insane Guidelines
Using jump trills of any kind that lead to a hand is discouraged. Unless a beatmap uses jacks frequently or other complex patterns frequently, this pattern is difficult to hit at a moderate to high speed. For 4K, this creates a mini jack that drastically increases the pattern difficulty. If a hand is really necessary, delete the note before it that would create the mini jack.

umm I know It's for guidelines but here something weird a bit.
If we restrict it, all the jump trills will have the same type to avoid a jack. and especially for 4K, the limit of SR is depends on the BPM. so most of them are ended as Insane difficulty. so most 4K mapsets will be same place on the ground If we mind it. so yeah It will make a huge change in the future. so before making the final decision for insane guidelines, need to think the past. what I'm trying to say is that the thing has never been considered in the past of osu!mania history since the communities are satisfied about the freedom of pattern choices. And yeah still not really have a negative impact because It's preferentially judged by BNs in the process of the ranked. so I wonder it should be on here.
If this is really needed, It should be applied for hard difficulty or when insane diff is not a last level of the mapsets(like ENH"I"I), should be.
so It has to make an exception case like this in my opinion.

anyways, all the rest are fine to me. good work!
MEGAtive

Arzenvald wrote:

5 consecutive 1/4 notes
Is this mean we can put 5 1/4 on faster bpm song? which certainly not cool, Or is there any certain bpm threshold for this?
I remember reading the preface of difficulty-specifics saying guidelines are using BPM180 as the standard and anything faster/slower should use it reasonably.

EDIT: Yea it's definitely up there. I just checked.
AchsanLovers

Kawawa wrote:

If this is really needed, It should be applied for hard difficulty or when insane diff is not a last level of the mapsets(like ENH"I"I), should be.
so It has to make an exception case like this in my opinion.
yea, im agree with kawawa
Full Tablet
Beatmaps can only use 4 to 9 keys. Anything else is not properly supported.
What does "not properly supported" mean? Other key modes play properly in the game client.

Long notes that are held simultaneously must be released at the same time. Long notes with multiple release timings require coordination that players of this level cannot reasonably handle.
Forbidding this pattern for Easy and Normal seems a bit excessive to me. If there is a progression in difficulty from Easy -> Normal -> Hard, players should attain the muscle memory required for basic LN patterns before they encounter some more advanced LN patterns in Hard maps.

I think this should be a guideline for Easy instead, and restricted only to LNs that are released with the same hand. For normal, make it a guideline that restricts different release times if they are in fingers that are next to each other. Also, for both difficulties, there should be a minimum of time between one release and the next one (about 2 seconds for Easy and 1 second for Normal sounds fine to me).
Shoegazer
Split jumptrill: A jumptrill which uses both hands at the same time to perform correctly - i.e two one-handed trills.
A [12][34] jumptrill also requires both hands to perform correctly. You might want to define it as a jumptrill which requires two one-hand trill motions to perform correctly instead.

The term "one-hand trill" should also be added if you have "split jumptrill" in the glossary.
Every note should correlate to a sound present in the music.
The guideline did not state whether the note-to-sound correlation must be consistent (barring exceptional circumstances). This is particularly important to distinguish because there are many maps that are not necessarily consistent in its layering in the ranked section, yet layering consistency is something that is valued in many mapping circles. I mentioned this before in the previous draft, but I didn't receive a response.

There are also certain notes that technically go to multiple sounds at the same time, like a LN where the LN head is justified by a hi-hat and the LN aspect of the note being justified by a syllable (vocals). Certain moments of the song with no percussion but vocals will not have any note in that moment of the map. How do you deal with cases like these, where a certain type of note (not but the note itself) correlates with a sound?

There are also some edge cases where notes do not strictly correlate with the start and end of a sound should be added because of the properties of the sounds that are supposed to be captured, but that's a different discussion altogether. I wouldn't mind discussing about this further with the people at hand.

Exceptions in favor of simplification should be made when following the music completely accurately...

would be unreasonable to play with full accuracy.
would be effectively represented with a consistent, predictable snap
would cause an inappropriate difficulty spike.
- Almost all instances where a section is unreasonable to play if a map were to capture a rhythm with maximal accuracy are due to inappropriate difficulty spikes in terms to either physical demand or rhythmic complexity. "would be unreasonable to play with full accuracy" is redundant and should be removed.

- What does this mean for instances like using a 1/6 stream for a 1/8 rhythm? A 1/4 non-LN stream (or even 1/2 non-LN stream) to capture a 1/8 rhythm has been permitted before, and a 1/6 stream is typically more effective in representing a 1/8 rhythm than a 1/4 stream in terms of general intensity. There are also cases like 1/16 swing that have to be considered.

- "Difficulty spike" needs to be defined further. Are you referring to difficulty spikes in terms of density, or can it refer to difficulty spikes in timing difficulty? Maximally accurate guitar solos are notorious for being very difficult to time for KB play, and they would be considered "spiky" despite not being too much harder than a straight 1/4 or a 1/6 stream.

Long notes should not have release timings that use different timing snaps. For example, long note releases should not have a 1/3 snap on the end of it and a 1/4 snap at the end of another.
The rule seems pointless without giving a justification for why this rule is instated. The only reason that I can think of is that you would like releases to be justified by the "release" aspect of a note, but many many maps already violate this rule if you were to care about releases that stringently.

There are also occasions where releases on different snaps would be justified based on where the notes end in the song.

Difficulty-specific guidelines:
If Normal is the lowest difficulty of a beatmap set, it has to abide by these additional rules:

Consecutive 1/4 snaps may only be two notes long. More dense rhythms are not appropriate for the lowest difficulty of a beatmap set.
This is extremely important (and I'm not sure if this is overlooked), but consecutive 1/4 snaps should be three notes long at least. In a map with mostly 1/2 rhythms, a 1/4 gallop (2 consecutive 1/4s) requires omitting a potential 1/2 note, which would cause timing difficulties (especially for new players) and major issues if you were to layer a certain sound consistently.
Jack usage is discouraged. These can be too physically demanding for players of this level. Minijacks with ample rest time in between may be acceptable.
You should specify that you are referring to 1/4 jack usage. 1/2 jacks (90 BPM 1/4s in this example) are usually far more bearable to players who are just approaching Hard difficulties.

If Hard is the lowest difficulty of a beatmap set, it has to abide by these additional guidelines:

Chords within streams are discouraged. The density of these patterns is often too difficult for players of this skill level, especially when used on the lowest difficulty of a beatmap set. However, using a chord at the beginning/end of a stream is fine.
I believe that giving a set frequency for the number of doubles that can be used per beat would be better here. Doubles every 2/1 seems like a more reasonable cutoff than "only a chord in the beginning and/or the end".
Avoid 1/8 streams lasting longer than 4 beats. Streams of this type of snap are hard to execute and should only be used if the song warrants it.
Is this only referring to non-rolly streams? There are several maps that have very long 1/8 rolls that are permitted into the ranked section (e.g. Blue Dragon). If you're referring to non-rolly streams, do include that.

Using jump trills of any kind that lead to a hand is discouraged. Unless a beatmap uses jacks frequently or other complex patterns frequently, this pattern is difficult to hit at a moderate to high speed. For 4K, this creates a mini jack that drastically increases the pattern difficulty. If a hand is really necessary, delete the note before it that would create the mini jack.
This does not align with the Hard guideline where 1/4 minijacks are practically not perimtted (barring minijacks that are very infrequent). A jumptrill is a very simple motion, and a minijack into a triple is not particularly difficult. I don't mind this rule personally, but it seems a bit odd to disallow minijacks like that when those patterns are only hard when they appear frequently.

Avoid more than 5 consecutive split jump trills. These are more physically exhausting than other patterns of similar density.
This should be rephrased as "5-note split jumptrills". That said, I think that 9-note split jumptrills should be the maximum instead of 5. This one is a bit more contentious, though.

Other things that I didn't list I didn't think they were issues, or I've never tried experimenting with (e.g. SVs).

Apologies if this was listed anywhere, but what is the goal of the new ranking criteria? I understand that the ranking criteria is supposed to be something like "minimum requirements" for maps, and the updated ranking criteria seems to have more specific rules on the minimum requirements for specific difficulties, but is there any other motive behind the updated ranking criteria?

For instance, there was a discussion on whether dump charting/mapping should be permitted, but there are several rules that go against this (e.g. the general LN length guideline and how each note must correlate to specific sounds). I'm personally ambivalent on the idea of allowing dumps, but I know that a fair section of the community is quite accepting of dump charting.

Besides that, are there any other goals or motives that the updated ranking criteria is set to accomplish? Again, I apologise if this has been mentioned before, but I haven't been in tune with the ranked section community or the ranking criteria because the circles that I'm in don't focus on the ranked section to begin with.
MEGAtive

Kawawa wrote:

umm I know It's for guidelines but here something weird a bit.
If we restrict it, all the jump trills will have the same type to avoid a jack. and especially for 4K, the limit of SR is depends on the BPM. so most of them are ended as Insane difficulty. so most 4K mapsets will be same place on the ground If we mind it. so yeah It will make a huge change in the future. so before making the final decision for insane guidelines, need to think the past. what I'm trying to say is that the thing has never been considered in the past of osu!mania history since the communities are satisfied about the freedom of pattern choices. And yeah still not really have a negative impact because It's preferentially judged by BNs in the process of the ranked. so I wonder it should be on here.
If this is really needed, It should be applied for hard difficulty or when insane diff is not a last level of the mapsets(like ENH"I"I), should be.
so It has to make an exception case like this in my opinion.

anyways, all the rest are fine to me. good work!
You could omit the second last jump as single and still ending the Jumptrill with triples, though. The pattern itself is fine but as far as I can see this kind of patterning are abused to pump up SRs which is not very good either. And speaking of Jumptrill variation, 90% Jumptrills I see up until now are full jumptrill ended with triples so I guess this guideline actually enforces Variation (?) Yea it wasn't a big deal but it may improve a map presentation quite a bit.

Shoegazer wrote:

This is extremely important (and I'm not sure if this is overlooked), but consecutive 1/4 snaps should be three notes long at least. In a map with mostly 1/2 rhythms, a 1/4 gallop (2 consecutive 1/4s) requires omitting a potential 1/2 note, which would cause timing difficulties (especially for new players) and major issues if you were to layer a certain sound consistently.
This is also my concern about this rule as gallops doesn't simplify a triplets. But I also thought that if Normal are using 1/4s triplets, wouldn't it be similar to Hard?
Honestly I never see an "Insane" map using chords mid-1/8 stream in ranked lul, even when the 1/8 streams are in roll. But my thought lies within the "Using a chord at the beginning or end of a stream is fine". You didn't specify what chord are acceptable for the stream opening and ending. Using triples at the start would made stuff feels awkward while using triples at the ending is pretty hard to execute. Or heck, you can still use quads for the opening and ending (which would be the worst case scenario but I think people aren't that dumb).

Yea, just define what chord are acceptable for the opening and ending portion of the stream.
Kawawa

MEGAtive wrote:

Kawawa wrote:

umm I know It's for guidelines but here something weird a bit.
If we restrict it, all the jump trills will have the same type to avoid a jack. and especially for 4K, the limit of SR is depends on the BPM. so most of them are ended as Insane difficulty. so most 4K mapsets will be same place on the ground If we mind it. so yeah It will make a huge change in the future. so before making the final decision for insane guidelines, need to think the past. what I'm trying to say is that the thing has never been considered in the past of osu!mania history since the communities are satisfied about the freedom of pattern choices. And yeah still not really have a negative impact because It's preferentially judged by BNs in the process of the ranked. so I wonder it should be on here.
If this is really needed, It should be applied for hard difficulty or when insane diff is not a last level of the mapsets(like ENH"I"I), should be.
so It has to make an exception case like this in my opinion. anyways, all the rest are fine to me. good work!
You could omit the second last jump as single and still ending the Jumptrill with triples, though. The pattern itself is fine but as far as I can see this kind of patterning are abused to pump up SRs which is not very good either. And speaking of Jumptrill variation, 90% Jumptrills I see up until now are full jumptrill ended with triples so I guess this guideline actually enforces Variation (?) Yea it wasn't a big deal but it may improve a map presentation quite a bit.

If It causes a potential point of making a mindblock, It does not have to be defined exactly as what I mentioned eariler because judgements about reasonable points are performed by BNs. but If It's included only that meaning what you said, I agree. It's yeah make sense. but well, perhaps because of the difference in the interpretation of the sentences. so yeah in case
Tailsdk

MEGAtive wrote:

Honestly I never see an "Insane" map using chords mid-1/8 stream in ranked lul, even when the 1/8 streams are in roll. But my thought lies within the "Using a chord at the beginning or end of a stream is fine". You didn't specify what chord are acceptable for the stream opening and ending. Using triples at the start would made stuff feels awkward while using triples at the ending is pretty hard to execute. Or heck, you can still use quads for the opening and ending (which would be the worst case scenario but I think people aren't that dumb).

Yea, just define what chord are acceptable for the opening and ending portion of the stream.
I think just talking about insane as 1 difficulty can be tough since a 4.8 star is still an insane difficulty. I personally think it should be split up into a higher insane and lower insane araound 4.4 or 4.5 since its a pretty big gap in between
show more
Please sign in to reply.

New reply