Guest
home
beatmaps
rankings
community
store
help
forum
  1. Forums
  2. Beatmaps
  3. Mapping Discussion
  4. Ranking Criteria
  5. Finalized/Denied Amendments

[Rule Change] Star difficulty change

posted
Total Posts
85
show more
Kodora
Kodora

Loctav wrote:

Let me purpose a proper spread rule according to the new star rating system:

The mapset must have a well-designed spread of difficulties, containing at least an Easy or a Normal difficulty. This is so that players of all levels of experience are able to enjoy maps of the songs they love.
The difficulties in the mapset must be in a consecutive order. The order can be seen in the chart below. If your mapset has two difficulties, one of them cannot be an Insane or Expert.
The difficulty level of Taiko-specific difficulties must also follow a well-designed spread and might contain an Hard/Insane only, if there are standard difficulties present.
The difficulty spread is determined by the map's star rating. A map falls under a certain difficulty when having a specific star rating:
  1. Below 1.5: Easy
  2. Below 2.25: Normal
  3. Below 3.75: Hard
  4. Below 5.25: Insane
  5. Above 5.25: Expert
I think we can use a bit higher value for Normal diffs (2.50 stars would be good imo - that should work for high-BPM songs or song with tricky/weird rhythm with possible rule that Normal diff should be below 2.25 stars for NH/NHI/NHIE spread) and for Hard diffs (4.00 would be great for same reason).

Also i think it would be nice idea to add Ephemeral's suggestion from previous thread as part of new rule:

Ephemeral wrote:

Acceptable spreads are NHI, ENH, or NH. No other spreads are allowed as a minimum requirement.
What do you think?
Loctav
Popcorn Fairy
4,496 posts
Joined February 2009
Loctav
2.5 ends up in it labeled as Hard. The values are taken from the star calculation code. 2.25 is the bare maximum.
Ephemeral's suggestion is pretty much covered with my rule purposal. (If your mapset has two difficulties, one of them cannot be an Insane or Expert.)
Kayano
1,286 posts
Joined July 2011
Topic Starter
Kayano
then what about the rules of CtB? Almost all CtB diffs (even a Cup=Easy) are above 3.75 stars so how can we judge the difficulty in quantitative way
Kodora
Kodora

Loctav wrote:

2.5 ends up in it labeled as Hard. The values are taken from the star calculation code. 2.25 is the bare maximum.
Like i said above, although new stars are much more accurate, they aren't work 100% perfectly - let's take some examples:

https://osu.ppy.sh/b/90517 - a high-bpm map with pretty weird rhythm, but normal diffs feels appropriate as for me

https://osu.ppy.sh/b/147986 - same goes here, 202 bpm song with tricky rhythm

https://osu.ppy.sh/b/177863 - same here, stuff like goreshit isn't really can be easier, and i honestly feel this normal diff pretty much appropriate for such song and with such difficulty spread)

https://osu.ppy.sh/b/168149 - This song just have "weird" rhythm, but this Normal diff spreads very well together with such Easy & Hard

https://osu.ppy.sh/b/153379 - Same goes here, with such E N H spread such 2.41 stars normal works perfectly

Yes, i can agree that all mentiored above diffs can't be used as easiest diffs in mapset, but they works pretty much fine as exceptions for such songs with appropriate spreads - that's why i suggest to leave 2.25 stars as minimum required for mapsets where Normal diff used as easiest diff in mapset, and leave 2.50 as exception only for sets with proper easiest difficulty.

Loctav wrote:

Ephemeral's suggestion is pretty much covered with my rule purposal. (If your mapset has two difficulties, one of them cannot be an Insane or Expert.)
Sadly this rule propose would allow inappropriate spreads like EN or EHI, which are badly balanced in general. I believe that mentioring minimum required acceptable spreads directly in this rule would also help newcomer mappers understand it properly. :)
Loctav
Popcorn Fairy
4,496 posts
Joined February 2009
Loctav
How would they allow that? It says, stuff must be consecutively. EHI is not consecutive.
Most Normals are too hard nowadays and I wouldn't mind if Normals would finally be Normals again.
I am sick of exceptions and this star rating finally offers an opportunity to draw a clear and fair border in this spread rule, without treating anyone unfairly.
Wreath
435 posts
Joined February 2012
Wreath

Loctav wrote:

How would they allow that? It says, stuff must be consecutively. EHI is not consecutive.
Most Normals are too hard nowadays and I wouldn't mind if Normals would finally be Normals again.
I am sick of exceptions and this star rating finally offers an opportunity to draw a clear and fair border in this spread rule, without treating anyone unfairly.
So you mean E and N must be the star rates what the criteria says even if the diffs are not hard to play?
Oh gosh how weird.

The criteria is not that fair at least to me.
Why don't you decide important things like that by voting from all players?
those
3,136 posts
Joined November 2010
those
If an exception has to be made, then it's not a problem with the spread rule; rather, it's a problem with the difficulty calculation formula.
BeatofIke
4,166 posts
Joined September 2011
BeatofIke
Just a curious question, do any ranked/approved mapset even exceed 7.1 stars? LOL
Come to think of it, I don't see any ranked mapset over 7.1 stars (unless you enable DT mod).
What's the point of a 10 star system when no maps can even exceed 7.1? :?
Vuelo Eluko
5,484 posts
Joined October 2013
Vuelo Eluko

BeatofIke wrote:

Just a curious question, do any ranked/approved mapset even exceed 7.1 stars? LOL
Come to think of it, I don't see any ranked mapset over 7.1 stars (unless you enable DT mod).
What's the point of a 10 star system when no maps can even exceed 7.1? :?
future growth.. and they can they just havent yet, take Tenshi for example, if the BAT's open their minds to it a little it could get ranked and set a precedent.

why should the star system cap out at the hardest current ranked maps? they're going to get harder eventually.
captin1
osu! Alumni
2,595 posts
Joined February 2011
captin1
To return a bit to the main topic, the best way to judge if a map is suitable as a lowest diff is still actually looking at the map. And in my opinion, any star rating cap for a lowest diff should be lower, rather than higher. Putting it at the highest peak of normal will result in some lowest diffs that are mostly unsuitable as the lowest difficulty in a mapset.

To give examples from my own mapsets, I certainly would not want this map or this map as the lowest difficulty in the mapset, as they are far too hard for a new player but still good as a "Normal" diff.
Kodora
Kodora

Loctav wrote:

How would they allow that? It says, stuff must be consecutively. EHI is not consecutive.
It still allows EN to be ranked while it is not allowed anymore. Making this wording more detailed will just help to avoid any confusion.

Loctav wrote:

Most Normals are too hard nowadays and I wouldn't mind if Normals would finally be Normals again.
I am sick of exceptions and this star rating finally offers an opportunity to draw a clear and fair border in this spread rule, without treating anyone unfairly.
Very subjective, and i'm disagree with this. Most of modern Easy/Normal diffs looks much more smoother, easier and beginner-friendly than older maps, and i honestly think that whole mapping did a great progress here. As proof to my words compare new star rating of Normal diffs at newest maps (2011-2014 mainly will be below 2.25, also newest ones would have in general less stars) and oldest ones (2008-2010 - some of them would be even higher than 3.0) - i'm not even talking that older easiest diffs have more choppy rhythm and quite weird patterns sometimes.

captin1 wrote:

To return a bit to the main topic, the best way to judge if a map is suitable as a lowest diff is still actually looking at the map. And in my opinion, any star rating cap for a lowest diff should be lower, rather than higher. Putting it at the highest peak of normal will result in some lowest diffs that are mostly unsuitable as the lowest difficulty in a mapset.
Yeah, totally agree - that's why i think that we need to have lower star rating for Normal diffs what are going to be easiest diff in mapset. Perhaps maybe even lower than 2.25? Some 2.25 Normals might be very tricky sometimes (for example my own Normal - although it is only 2.08 stars i don't think that it would be appropriate as easiest diff in this mapset).
Last edited by Kodora , edited 1 time in total.
Sieg
3,981 posts
Joined February 2012
Sieg

Loctav wrote:

Most Normals are too hard nowadays and I wouldn't mind if Normals would finally be Normals again.
yep, it would nice to have even 1.8 as lowest.
haha5957
1,433 posts
Joined July 2010
haha5957
Why dont you all start thinking about that every song has different BPM? like, I don't wanna image 1.1 star diff freedom dive because it's going to be fucking fuol of 1/1 or even 2/1(not 1/2!) sliders.

A soft rule saying "A mapset needs to have lowest difficulty that can reasonably accepted as easy, or normal depending on the music"should be more than enough.

A beginner needs some map to play? Relaxing music will have something like 2.6 star rating as its highest and maybe even 0.8 for lowest difficulty.

I'm not saying that we should follow other rhythm games, but take a look at jubeat - Evans has 6/8/10 diff spread, macuilxochitl 7/8/10, albida 5/7/10. However, for others, there are songs that have diff spread of 1/3/5, 1/4/6, 2/4/6, 3/6/8.,,

See the point?

fast and crazy songs look really retarded with very low star rating. Relaxing, or even like, most of anime OP easy/normal diffs will satisfy beginners, and it doesnt even take that long to get enough skill to play normal+~Hard- diff.

I'm not saying something like "ok lets allow insane as an lowest difficulty if the music has 250+ bpm!". I'm saying that trying to set way too low star rating as a defined borderline for easiest diff would be weird.

How about setting a defined borderline for hardest diff! hardest difficulty should be 3.75 star or higher?
sounds fucking retarded, right? because relaxing songs are like, nearly impossible to have high difficulty without excessive overmapping. although it's not exactly the same, the point is still there,
Sieg
3,981 posts
Joined February 2012
Sieg

haha5957 wrote:

Why dont you all start thinking about that every song has different BPM? like, I don't wanna image 1.1 star diff freedom dive because it's going to be fucking fuol of 1/1 or even 2/1(not 1/2!) sliders.
What's wrong with easiest difficulty in mapset with full of 1/1 or even 2/1?
haha5957
1,433 posts
Joined July 2010
haha5957
you can make easy or normal diff actually feel like playing the song if you are playing relaxing map. I mean, it doesnt even need to be relaxing map.

take a look at easiest diff on http://osu.ppy.sh/s/54016

It pretty much feels like playng the music even with minimum beat mapped. how? because the BPM is 96.

do you really think you can feel a rhythm and joy with map that is full of 2/1 going on when the music is superfast such as dance number and freedom dive?

I mean, even the answer is yes for that, do you really think that rule needs to force that?
Sieg
3,981 posts
Joined February 2012
Sieg

haha5957 wrote:

do you really think you can feel a rhythm and joy with map that is full of 2/1 going on when the music is superfast such as dance number and freedom dive?

I mean, even the answer is yes for that, do you really think that rule needs to force that?
for sure

some examples for you
https://osu.ppy.sh/b/140579
https://osu.ppy.sh/b/169520
https://osu.ppy.sh/b/243551
https://osu.ppy.sh/b/273159
https://osu.ppy.sh/b/271581
Kodora
Kodora
Agreed with haha5957

Making "iron" limitations for star rating would ends up in poorly mapped easiest diffs with only bland 2/1 or 1/1 spamming.

Rhythm matters even on easiest diffs - beginners should enjoy the song, not enjoy constant 2/1 or 1/1 hitting
DakeDekaane
osu! Alumni
3,500 posts
Joined February 2012
DakeDekaane
I thought we were discussing the max star rating required for a N as the lowest diff.

I think 2-2.1 stars is enough to have not a dull normal but also not too difficult. If you need more freedom, then add an Easy, I really doubt they'd go higher than 2 stars even with complex rhythm.
ZenoDiac
117 posts
Joined August 2011
ZenoDiac
Yes, but seriously though getting mapper to make easy beatmaps on a high bmp track like DJ Sharpnel - StrangeProgram really just calls for a useless map just to make it fit into the system.

A better solution would be to limit the star rating accordingly to a set bpm.
haha5957
1,433 posts
Joined July 2010
haha5957

DakeDekaane wrote:

I thought we were discussing the max star rating required for a N as the lowest diff.

I think 2-2.1 stars is enough to have not a dull normal but also not too difficult. If you need more freedom, then add an Easy, I really doubt they'd go higher than 2 stars even with complex rhythm.
I'm saying this because "Adding an easy" could be pain in the ass for high BPM songs(and I'm pretty sure that you know BPM is not the only factor)

like, I thought the point of "map must having easy/normal difficulty" was to have mapper actually map "minimum amount of beat to feel a music", so it can possibly be played by beginners.

Sorry for repeating myself, but imagine we having rule something like "Map must have a difficulty with star rating of 3.75+ for advanced player's game experiance". Wouldn't it be stupid and weird? You can still add bunch of weird feeling triplet,stream,jumps to boost difficulty, but you just don't want to do it because it won't fit that good.

But you don't blame relaxing musics having 4.0(I mean, the old star rating which is more like 2.3 nowdays) difficulty as their hardest difficulty, and now why are you blaming a crazy music that has 2.4 star rating as their easiest difficulty?

I understand Sieg's example for "Extreme music, easy difficulty" but I still don't get the point. Should it be really forced?

I'm not saying that it "SHOULD NOT" have the easy difficulty. I'm just saying that it does not need to be forced.
Zare
1,512 posts
Joined March 2012
Zare
idk guys how about we just use our brains and say "a mapset needs something that doesn't look like a complex, unreadable, impossible map for complete beginners"
because that's what the rule is about essentially
incidentally, the new star rating draws a line around there, judging somewhat passable-seeming maps (for newcomers) as Normals, and anything above that is hard

having that said, i actually, god forbid, agree with loctav on just using the new star rating defined limitations for ranking criteria
heck I expected the elitists here to try and enforce an actual Easy diff now that it's actually possible to map Easy diffs without overshooting the star rating just by placing 2 more objects, so just needing a Normal is enough freedom for mappers
Garven
osu! Alumni
5,857 posts
Joined January 2010
Garven
The point of the low star rating requirement is because high level players can play easier maps. It doesn't go the other way around. End of story. If you want to try writing a new chapter, make a thread about it. This isn't the right place for that line of discussion.

Also: keep the profanity down haha5957.
WingSilent
2,982 posts
Joined August 2012
WingSilent
Now, having a 10★ beatmap would mean barely impossible to play.. alright.
Loctav
Popcorn Fairy
4,496 posts
Joined February 2009
Loctav
Have you ever considered that higher BPM is simply harder for beginners?
Anyways, according to Tom, 10 star maps are basically almost impossible, if not impossible to pass (corrrct me if I am wrong, Tom :( )
Nevertheless, I will ask someone to revise the wording of my rule, but we will not move away from the concept of it. You either have the choice to help in wording this or leave this thread. The rule might be readjusted according to star rating behavior, in case it gets slightly changed, but the rule self will stay.

Bubbled. Fix my wording.

Edit: oh yes, what captin1 said seems reasonable. I wonder if it might be suitable to have the easiest difficulty be at least below 2.0 stars. (So if you map a normal only, it must be an easier normal - if you add an Easy, your normal can be harder)

And don't forget: if a difficulty is too hard as Normal or too easy as Hard (I mean: where a Normal is labeled as Hard, even tho it plays like a normal), it's clearly a flaw in the difficulty calculation and can be adjusted by Tom. Whenever a difficulty seems to be incorrectly labeled, it migut offer room for star rating calculation improvement. But for this, we have a thread in General development, opened by Tom. This is the right place to post for this. (I basically repeated those, but he got blatantly ignored)
Flower
1,723 posts
Joined August 2011
Flower
The problem is that old star system was weird... Example is the oldest Can't Defeat Airman (Insane difficulty, Normal icon in old stars). 3 old stars can still be hard for new players, as long as the mapper puts lots of non-break pauses but sometimes intense patterns. Deciding if the lowest diff is easy enough for the beginner is, and will be, a case-to-case problem (if anyone remembers that a lowest diff must be passed by an imouto-chan).
Anyway, I can accept the new limit be 2.5-3, or 2, hence it is clear and easy to be handled. Hope this get passed in a second so I can get my map ranked...
BeatofIke
4,166 posts
Joined September 2011
BeatofIke
The new system seems better, but the star rating standards are TOO HIGH!
Image material 7 stars. REALLY?

I could imagine seeing what a 10 star map would look like. Is it even possible? >.<
Cherry Blossom
3,185 posts
Joined October 2011
Cherry Blossom

BeatofIke wrote:

I could imagine seeing what a 10 star map would look like. Is it even possible? >.<


That's above 10 stars lol

SPOILER

Edit, you can't imagine this

SPOILER
Last edited by Cherry Blossom , edited 1 time in total.
Zare
1,512 posts
Joined March 2012
Zare
@BoI
No
That's the point. By having 10 star maps being pretty much unplayable we have an insurance that ranked maps are not gonna exceed our soft cap of 10 stars at any time.
At the same time it funtions as a reference for mappers and players

When we had 5 stars, we already had playable maps above the limit. Why create a soft cap that has already been broken? (and yes we could probably use 8 stars or something and have the same effect, but that wouldn't look as good :v)
Vuelo Eluko
5,484 posts
Joined October 2013
Vuelo Eluko

Zare wrote:

That's the point. By having 10 star maps being pretty much unplayable we have an insurance that ranked maps are not gonna exceed our soft cap of 10 stars at any time. (in the near future)
people will eventually reach that level.
Zare
1,512 posts
Joined March 2012
Zare
No.
people won't reach a level where they can play 330 BPM 1/4 fullscreen pentagons
Kodora
Kodora

Zare wrote:

No.
people won't reach a level where they can play 310 BPM 1/4 fullscreen pentagons
lol are you sure?

no kidding actually, it really may happens
WingSilent
2,982 posts
Joined August 2012
WingSilent

Cherry Blossom wrote:

you can't imagine this

SPOILER
Well, hum.. i'd consider this as a bug. :o
Loctav
Popcorn Fairy
4,496 posts
Joined February 2009
Loctav
Scheduled discussion end: 23rd June 2014
Chewin
osu! Alumni
4,082 posts
Joined December 2010
Chewin
Mh.. I was thinking something different, if it is possible to apply lol.
Deciding the diff settings of the mapset starting from the BPM of the song and hence derive the star rating from the used setting according to a formula or to a mathematic average. Well, it's a bit hard to explain it so this should make you get the idea:

BPM 100~140

Easy: The map must have only HP: 1~3 OD: 1~3 AR:1~3 CS:1~2 (If they use like 1/1/1/1 the system should automatically set 0.5 star rating, if 2/2/2/2 should set 1.00, if 3/3/3/2 should set 1.50, if 1/2/2/1 should be 0.75 and so on)
Normal: The map must have only HP:3~5 OD: 3~5 AR: 3~5 CS:2~3 (If they use like 3/3/3/2 the system should automatically set 2.00 and so on like I wrote before and being consistent following the same formula)

BPM: 150~190

Easy: The map must have only HP: 2~3 OD: 2~3 AR:2~3 CS: 1~2 (for all low settings as the easy before still 0.5 and so on, I mean this because higher bpms require higher settings to make it easier or it would be hard. For example: a Normal at 190 BPM is easier with AR 5 than AR 3~4)

This is the same for H/I/X till getting the 10 star rating when we can have kinda 10 star for all max settings. (Still considering the used BPM).
Starting from this we can consider rankable only mapsets with ENH or EN or ENHI or NHI or NH or NHI.

Well, it should follow a "math" criteria, kinda.
It's not easy to explain the concept but I hope you got what I wanted to suggest to do.
Loctav
Popcorn Fairy
4,496 posts
Joined February 2009
Loctav
Shouldn't this all be represented by the star difficulty calculation? You still point up flaws in the star difficulty calculation and not in the rule self.
Chewin
osu! Alumni
4,082 posts
Joined December 2010
Chewin
This would include the rule self if applied as I said here:

Starting from this we can consider rankable only mapsets with ENH or EN or ENHI or NHI or NH or NHI.
w/o considering the number of star rating since it has been already considered rankable because of the "logical" star rating formula
Kodora
Kodora

Chewin wrote:

CS:1~2
Isn't 1 is unrankable?

Also i'd personally never suggest making such iron limitations at least for settings - i dislike using huge circle sizes for example.
Kazuya
894 posts
Joined May 2013
Kazuya
Sorry, my english is = 0
----------------------

It's only ideas.
Easy:

AR: 2-3
OD: 1-3
HP: 1-3
CS: 2 or 3

90BPM - 140 BPM :
Slider Velocity: Max 1.20
Distance snap: Max 1.0

141BPM - 170 BPM :
Slider velocity: max 1.00
Distance Snap: max 1.0

170BPM - 200 BPM :
Slider velocity: max 0.90
Distance Snap: max 1.0

201BPM - 250 BPM :
Slider velocity: max 0.70
Distance Snap: max 1.2 or 1.4 ?


250 BPM -
Half impossible (?)

Normal:

AR: 4 or 5
OD: 4 or 5
HP: 4 or 5
CS: only 3

90BPM - 140 BPM :
Slider Velocity: Max 1.40
Distance Snap: Max 1.2

141BPM - 170 BPM :
Slider velocity: max 1.20
Distance Snap: max 1.0

170BPM - 200 BPM :
Slider velocity: max 1.00
Distance Snap: max 1.0

201BPM - 250 BPM :
Slider velocity: max 0.80
Distance Snap: max 1.4

250 BPM -
Half impossible (?)

----------------------------------------------
if the mapset only (N/H/I):
Normal (Easy?):

AR: 4 or 5
OD: 3 or 4
HP: 3 or 4
CS: 2 or 3

90BPM - 140 BPM :
Slider velocity: max 1.20
Distance Snap: max 1.0

141BPM - 170 BPM :
Slider velocity: max 0.90
Distance Snap: max 1.2

170BPM - 200 BPM :
Slider velocity: max 0.80
Distance Snap: max 1.2

201BPM - 250 BPM :
Slider velocity: max 0.70
Distance Snap: max 1.2
/me runs.
Chewin
osu! Alumni
4,082 posts
Joined December 2010
Chewin

Kodora wrote:

Also i'd personally never suggest making such iron limitations at least for settings
They are the usual used settings anyways, Kodora. -3-"
DakeDekaane
osu! Alumni
3,500 posts
Joined February 2012
DakeDekaane
A single fixed number would work better than a lots of stuff imo, that way we keep things simple and not confusing.
show more
Please sign in to reply.

New reply

0
13 / 85