Guest
home
beatmaps
rankings
community
store
help
forum
  1. Forums
  2. osu!
  3. Development
  4. Announcements (Archived)

osu! Public Release b1596

posted
Total Posts
53
peppy
19,314 posts
Here since the beginning
Topic Starter
peppy
I don't feel like writing a long post here, so let's make this short. This new build fixes a few pending bugs (that were quite possibly introduced in the last build), and has a few minor improvements that most of you won't notice. The biggest addition is the new Difficulty Star system, which will eventually replace the current start difficulty algorithm. While it is still in testing stages, you can play with it via AiMod in the editor and post your feedback in this thread.

For more info on the difficulty system, you can read the wiki page (although the values on there are very outdated and have been tweaked a lot since that post).

If you have any adjustments you think should be made, please make a very clear case posting map(s) which you believe are falling in the wrong difficulty, along with the cause for this and your proposed fix. Note that you can copy-paste from the AiMod grid control.

The new difficulty system will likely be finalised before the end of the month as it is one of the prerequisites for a larger change which is soon to come. All feedback is appreciated.

I'll repeat for clarity: Difficulty Star calculations are not final. Some elements are still naive or need further adjustments. If you are going to criticise, please suggest alternatives and provide examples.

Full changelog:
SPOILER
(+) Special mode selection screen supports widescreen.
(+) Initial difficulty stars system implementation.
(+) BanchoBot gets a nice pink shade in in-game chat.
(+) AiMod selects related hitobjects where applicable.
(+) Added severity icons to AiMod.
(+) Added message about difficulty modding being in testing.
(+) Added AiMod difficulty rules: time between objects and streams.
(+) Added AiMod difficulty rule: spacing infringements.
(*) i265 (/me/np sent to self in PM).
(*) Slight reduction in osu!direct preview volume (ingame).
(*) Project cleanup and various fxcop rule fixes.
(*) Override spam check for admin users.
(*) Minor special mode select screen fixes.
(*) Increase maximum slider velocity to 3.6.
(*) General bancho optimisations and minor fixes (main).
(*) Further PrivilegeManager optimisations.
(*) Fixed textbox committing causing incorrect handling in some cases.
(*) Fixed song select jumping to weird places when it shouldn't.
(*) Fixed skip button not working properly after exiting the editor in design-mode (I304).
(*) Fixed osu!direct panel buttons not working correctly when many downloads are queued.
(*) Fixed multiplayer games starting prematurely with beatmaps with a leadin. (I306).
(*) Fixed mouse not reappearing early enough before break ends.
(*) Fixed i300 (Old URL for user links).
(*) Fixed i300 (Old URL for user links).
(*) Fixed i299 (half-second lag when switching chat tabs).
(*) Fixed i298 (Editor design tab breaks some storyboards).
(*) Fixed i294 (Global volume doesn't affect osu!direct preview).
(*) Fixed i292 (Starting multi match while someone is refreshing song).
(*) Fixed i289 (Highlight broken).
(*) Fixed i285 ("Jump to specific time" treats zeros in milliseconds incorrectly).
(*) Fixed i280 (Chat text moves after window resize).
(*) Fixed i273 (osu!direct + widescreen = cursor restricted).
(*) Fixed i272 (Widescreen + Taiko = graphic bug).
(*) Fixed i264 (Highlights don't save unless you press enter).
(*) Fixed i259 (Song filenames starting with numbers break the auto chat URL).
(*) Fixed i258 (Scoreboard shows ?? for #40).
(*) Fixed i252 (osu!direct alignment bug)
(*) Fixed i233 (Editor asks to save right after saving).
(*) Fixed hard crash in some rare incorrect skin.ini cases.
(*) Fixed for lead-in time on maps that take longer than normal to load.
(*) Fixed BAT members not automatically being admin.
(*) Fix threading bug with AiMod refreshing on map change.
(*) Fix out-of-range bug on map completion.
(*) Fix AIMod not closing when entering test mode.
(*) Fix #2 for i294 (osu!direct volume mismatch).
(*) Fix #2 for i214 (AltGr behaviour).
(*) Ensure osu! file associations are run during elevated privileges when installing.
(*) Changes to file escaping to allow better support for '$' character.
(*) Changed keyboard input handling to allow editor shortcuts to work while chat is being displayed.
(*) Bancho connection constant tweaks (improved connection reliability).
(*) AiMod variable adjustments.
(*) AiMod refresh.
(*) AiMod crash fix and minor modifications.
(*) AiMod bug fixes.
Kaya-_old
Kaya-_old
osu! don't work after update for me D:
System.NullReferenceException: В экземпляре объекта не задана ссылка на объект.
   в #1d.#Oh.#yJ()
   в System.Threading.ThreadHelper.ThreadStart_Context(Object state)
   в System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state)
   в System.Threading.ThreadHelper.ThreadStart()
(Also:where can I check when my profile will be unbanned ?)
peppy
19,314 posts
Here since the beginning
Topic Starter
peppy
@Uber1337: Should be fixed after you update again.
Kaya-_old
Kaya-_old

peppy wrote:

@Uber1337: Should be fixed after you update again.
Yay!
Works fine!
Mashley
2,042 posts
Joined September 2008
Mashley
Aimod is showing spinners as too close the previous spinner. Just thought I'd point that out~
Shinxyn
osu! Alumni
1,179 posts
Joined June 2008
Shinxyn
Stacks are now "difficulty" mistakes D:!
nvm, read it wrong :x
Last edited by Shinxyn , edited 1 time in total.
peppy
19,314 posts
Here since the beginning
Topic Starter
peppy

Shinxyn wrote:

Stacks are now "difficulty" mistakes D:!
You are misunderstanding the purpose of the difficulty tab. Try reading the opening post of this thread, and if you still don't understand it ask in chat.
Natteke
6,538 posts
Joined August 2009
Natteke
CtB seems to be bugged? If you start to play any song, the game just closes down
DeathxShinigami
1,176 posts
Joined November 2008
DeathxShinigami

Natteke desu wrote:

CtB seems to be bugged? If you start to play any song, the game just closes down
Confirmed. I don't even recieve the standard osu! crash message ethier. Might just be a windows bug?
hoozimajiget
45 posts
Joined December 2009
hoozimajiget

Natteke desu wrote:

CtB seems to be bugged? If you start to play any song, the game just closes down
Problem signature:
Problem Event Name: APPCRASH
Application Name: osu!.exe
Application Version: 1.3.3.7
Application Timestamp: 4bcfc690
Fault Module Name: KERNELBASE.dll
Fault Module Version: 6.1.7600.16385
Fault Module Timestamp: 4a5bdaae
Exception Code: e053534f
Exception Offset: 00009617
OS Version: 6.1.7600.2.0.0.256.48
Locale ID: 4105

that is the windows error report i get when starting CtB (not sure if its needed or not)
mikhe
673 posts
Joined November 2008
mikhe

Natteke desu wrote:

CtB seems to be bugged? If you start to play any song, the game just closes down
Peppy.. should fix this quickly ^^


-What happen with this stars??? """" Will you put it??




-And look: (Screen resolution 1024x600)
SPOILER
Ekaru
3,617 posts
Joined July 2008
Ekaru
Just a few questions/suggestions.

1. Will just one violation result in an otherwise 100% Hard map being pushed to Expert due to a single jump that isn't too difficult in reality?

EX: http://osu.ppy.sh/s/6342 (I recommend checking without your headphones on)

2. Is part of the reason you guys came up with this not only to make the difficulty system better but also to maybe improve the quality/balance of ranked maps a bit? Just wonderin'.

3. Can you make it so that if a, say, 4-second-long spinner is 39 ms after a hit circle it doesn't trigger an Expert violation? Because that would be an exception to the 40-ms thing IMO.
EX: http://osu.ppy.sh/s/6542

4. I would consider making an Overall Difficulty of 8 or higher an Insane criteria. From my personal experience and from what I've seen, the jump from overall difficulty 7 to 8 is/was a hurdle for quite a few players. Doesn't really matter though, just my opinion.

Otherwise it's quite good and fair right now, I say. Good work!

EDIT: Except, once the spinner bug gets fixed, this bullshit will be considered a "Hard":

http://osu.ppy.sh/s/4080 (Check the Insane, it's quite crappy)

A 57-hit-object-long stream with all gaps being 1/4 apart at 150 BPM should maybe trigger an Insane, possibly Expert criteria?
Last edited by Ekaru , edited 1 time in total.
PedroGabriel
32 posts
Joined August 2009
PedroGabriel
when play ctb

osu! stopped working

Description:
Stopped working

Problem signature:
Problem Event Name: APPCRASH
Application Name: osu!. Exe
Application Version: 1.3.3.7
Application Timestamp: 4bcfc690
Fault Module Name: KERNELBASE.dll
Fault Module Version: 6.1.7600.16385
Fault Module Timestamp: 4a5bdbdf
Exception Code: e053534f
Exception Offset: 0000b727
OS Version: 6.1.7600.2.0.0.256.1
Identification of Location: 1046


options:
Close the program
debug the program

the same error here >.<
miochan99
1 post
Joined October 2009
miochan99
CtB crashes for me as well after the update. I don't play it often, so it doesn't really bother me, but it should be fixed.
aston_old
aston_old
^ confirmed CtB crash over 5 times in a row.

perhaps its the maximum slider velocity increase?
ztrot
osu! Alumni
4,378 posts
Joined February 2008
ztrot
The judgement for notes after spinners is a bit to harsh Imo, possibly make it a bit more lax? all around? Map http://osu.ppy.sh/s/15113
Example
00:24:231 Hard criteria: Note exists 329ms after spinner (less than 400ms). I really don't think that should be considered as hard, seeing as most easies use alot of spinners, and notes almost always come after the spinner on a 1/1 scale I know that is based on bpm but I thing the judgement is a bit to harsh atm.
Zekira
9,233 posts
Joined September 2008
Zekira
1. Will just one violation result in an otherwise 100% Hard map being pushed to Expert due to a single jump that isn't too difficult in reality?

EX: http://osu.ppy.sh/s/6342 (I recommend checking without your headphones on)
I'd just like to confirm something in relation to this... http://osu.ppy.sh/s/14151 -- [Oni INSAAAAAAAAAANE] has 14 difficulty notes that involve non-stack streams, ah and the really fast stacks weren't counted to be a difficulty note, while [Muzukashii HAAAAAAAARD] only involve difficulty notes that are actually stacks, and yet is still considered Insane while Oni is only Expert. Also, Oni has a 9 for an OD. It's... weird? I've been wondering what the cause of the erratic star ratings on these maps were, and I never expected it to be the stacks and streams. Clarifications please? I'm getting more and more confused by the minute
Lybydose
osu! Alumni
2,024 posts
Joined January 2009
Lybydose

Zekira wrote:

1. Will just one violation result in an otherwise 100% Hard map being pushed to Expert due to a single jump that isn't too difficult in reality?

EX: http://osu.ppy.sh/s/6342 (I recommend checking without your headphones on)
I'd just like to confirm something in relation to this... http://osu.ppy.sh/s/14151 -- [Oni INSAAAAAAAAAANE] has 14 difficulty notes that involve non-stack streams, ah and the really fast stacks weren't counted to be a difficulty note, while [Muzukashii HAAAAAAAARD] only involve difficulty notes that are actually stacks, and yet is still considered Insane while Oni is only Expert. Also, Oni has a 9 for an OD. It's... weird? I've been wondering what the cause of the erratic star ratings on these maps were, and I never expected it to be the stacks and streams. Clarifications please? I'm getting more and more confused by the minute
Expert is higher than Insane.
Zekira
9,233 posts
Joined September 2008
Zekira

Lybydose wrote:

Expert is higher than Insane.
Yeah I failed and figured that out just now, but it's still odd how Muzukashii is still getting rated higher

Meh w/e I'll see what I can do about it
Lybydose
osu! Alumni
2,024 posts
Joined January 2009
Lybydose
Spacing violations are too strict for a couple reasons.

1. The "rule" for spacing in a combo is entirely determined by the distance between the first two notes. So, if the 1 and 2 are spaced slightly further than normal, but 2-10 follow a consistent distance, this will show up as 8 spacing violations in AIMod. It should instead show up as just one violation.

Example: 1-2-3 are jumps in this combo, but the rest of it is normally spaced. However, AIMod gives this like 9 distance errors because everything after 3 is "too close"

SPOILER

This may not necessarily even occur due to jumps; it could also happen due to one note being like 1-2 "level 3 grids" different for alignment or symmetry purposes, and it ends up causing the entire combo to be "wrong"

2. A lot of people like to intentionally give streams smaller distance spacing, but not actually stack them. When this happens, it causes AIMod to pick up difficulty increases for something that is actually easier.

Example: This stream intentionally has smaller spacing to make it easier to hit. Unfortunately, this causes like a million AIMod errors.
SPOILER

Had I spaced these out normally, AIMod would rate this as a lower difficulty, despite being much harder:

SPOILER

3. Doesn't take song length into account. A song with 40 "distance errors" over 3 minutes would probably play differently from a song with 39 "distance errors" over 30 seconds.

4. Particularly harsh on "accelerated spacing", "decelerated spacing", and "ztrot spacing", despite these things usually being pretty easy to pick up on and hit (well, for Hard or higher difficulty anyway). These things will give lots of errors. It should probably result in a difficulty increase past normal, but currently it gives so many errors that it essentially forces the map into Expert if you do this a couple times.

In addition, "notes after spinner" is generally too harsh as well because it doesn't take into account the length of a spinner. A note 1 beat away from a spinner might be really easy to hit if the spinner is long (>1 bar for example), or really hard to hit if it's like an auto +2000 spinner.

Deathstreams should also be rated higher. For example, this only comes out to "Insane" rather than "Expert." Wat.

SPOILER

Extra difficulty on this map is also a good example.
arien666
6,290 posts
Joined May 2009
arien666
Err???
Expert diff?

Wow! I'm so happy that this diff appears XD
[Helps for osu!arcade?<]
Anyway, keep going pep :3



edit.
About New Star Rating System...
SPOILER
★ - Easy-Normal (Conditions that when hit force a map to be at least normal difficulty)
Has a spinner under 1.5 seconds in length.
Has notes less than 150ms apart. (lesser than/equal to 1 beat apart? (1/1 divisor))
Has notes less than 2 seconds after a spinner. (perhaps reduce to 2 beats)
Overall difficulty or HP drain is greater than 2
Slider velocity is greater than 1.00
Average gap between beat placement is equal to or smaller than 1 beat
Has combos with a greater note count than 6
I suggest about 'Beginner' diff for this system ;_;
Just for beginners imo :3...
[Maybe, I believe that you thought about this ;_;]
Sorry :3...
Zekira
9,233 posts
Joined September 2008
Zekira
Deathstreams should also be rated higher. For example, this only comes out to "Insane" rather than "Expert." Wat.
This.
[DR]
16 posts
Joined March 2009
[DR]
Just me or you cannot push 'F5' after importing a map in multiplayer as you normally could to refresh the score on the map if you had already played it before.
peppy
19,314 posts
Here since the beginning
Topic Starter
peppy

[DR] wrote:

Just me or you cannot push 'F5' after importing a map in multiplayer as you normally could to refresh the score on the map if you had already played it before.
Yes, I removed the ability to refresh when it is unnecessary. There is no reason to.

arken1015 wrote:

I suggest about 'Beginner' diff for this system ;_;
Just for beginners imo :3...
Sorry :3...
Easy ~= beginner. I don't see the need for an easier difficulty.

I am reading all the feedback about the star system but probably won't reply in this thread unless I disagree. Your feedback helps with changing variables to judge maps better :).
Last edited by peppy , edited 1 time in total.
jericho2442
osu! Alumni
1,918 posts
Joined March 2009
jericho2442
well i gotta say i was one who has not liked some idea's done lately, but this one i think is a VERY nice idea.

great job on coming up with this :) will help out alot of new mappers and also give us more long term mappers more of a hand.

one thing though. will this also blend into what bats look for? as in if the aimod comes up clean for a normal difficulty in all area's would a bat class that map as fine ?

anyway nice work on this idea :)

EDIT: only rule ive found so far what i would say is a lil too harsh is the "easy" spinners distance to the next notes. just feels abit far than needed to me
mm201
osu! Alumni
4,495 posts
Joined August 2008
mm201
I think it would be an improvement to rename the existing difficulty names down one. ie:
Easy → Beginner
Normal → Easy
Hard → Normal
Insane → Hard
Expert → Insane

re: Lybydose spacing: I think spacing for a combo should be determined by the "mode" spacing of the combo, ie. the spacing which is used the most often, rather than the spacing of the first two notes. There are also situations where I think multiple combos should be treated as one for spacing purposes.

I'd consider ztrot formations to be made entirely of jumps, so they should be treated as such. If the jumps aren't especially hard, they should be accepted into a lower difficulty margin. (but no lower than Hard)

And something I discussed with peppy in chat--kind of the mirror image of the streams thing: Sometimes, to create a more visually appealing pattern, the mapper will widen the spacings of 1/3s or 1/4s (at a sufficiently low tempo) so that the notes don't overlap but are right next to each other. This is normally done in a way which isn't too difficult, such as doubles or a note after a slider end. This is considered intuitive and acceptable by most and I use this all the time.

Fudge spacings should be tolerated for notes whose "correct" spacing is at least 1.5 circle-widths apart, and the tolerance should be within the margin of 0.75x to 1.33x of their usual spacing.

The slider speed restrictions aren't very kind towards "big" mapping--look at maps by me, m980, Larto, and a few others. The maps have large circles and a large screen presence, so the spacings can be wider and the slider speeds can be faster without raising the difficulty too much. A typical speed/difficulty breakdown for a map of mine with an average, 130ish BPM would be something like:
Easy: 1.40x
Normal: 1.80x
Hard/Insane: 2.20x
The sliders don't feel unnaturally fast in the resulting map.

I think that note overlap should be considered one unique spacing distance, and no discriminations should be made based on how widely spaced the overlapping notes are. (What constitutes overlap would of course vary based on circle size.)

Lastly, constant circle sizes for Easy/Normal/Hard is a very common technique, so I think that Size 4 circles should be acceptable for Easy difficulties. Mouse sensitivities notwithstanding, there's a bit of a proportionality between circle size and distance spacing/slider speed, so I think that this ratio matters more for difficulty than either spacing or circle size on their own.
arien666
6,290 posts
Joined May 2009
arien666

peppy wrote:

Easy ~= beginner. I don't see the need for an easier difficulty.
Then, I don't have any other things to counter this :3



BTW...
imo, we need 'Difficulty number'[Like Drummania]' due to this orz situation...

viewtopic.php?t=26849

Zekira wrote:

Uh, why the hell does Muzukashii have a higher star rating than Oni?
I can't suggest anymore about this but...
It would be helpful for osu!arcade imo...



[Seems quickly deleted? ;_;]
0_o
2,155 posts
Joined December 2008
0_o
-Personally I prefer the name Extreme instead of Expert. It conveys more difficulty than Insane (and Expert), and it also fits with the "theme" of the other difficulty names: "Expert" describes the skill level needed to play the map, while "Extreme" - like Easy, Normal, Hard and Insane - describes the difficulty of the map itself. Just my two cents :P

-Like some of the others have already mentioned I think there should be a bit more breathing space for "spacing infringements". Maps don't need to be 100% distance snapped to be intuitive, and it's not like having a circle a few grids off is going to add much difficulty to a map.

-Slider speed guidelines should be dependent on circle size - a given slider velocity will appear to be faster slower with large hitcircles than smaller ones.

I like where this system is going, hope it works out ^_^
Last edited by 0_o , edited 2 times in total.
mm201
osu! Alumni
4,495 posts
Joined August 2008
mm201

0_o wrote:

a given slider velocity will appear to be faster with large hitcircles than smaller ones.
Backwards. :P

Keeping proportions, a slider speed may be faster if paired with larger circles.
0_o
2,155 posts
Joined December 2008
0_o

MetalMario201 wrote:

0_o wrote:

a given slider velocity will appear to be faster with large hitcircles than smaller ones.
Backwards. :P

Keeping proportions, a slider speed may be faster if paired with larger circles.
The funny part is I spent a good 10 minutes trying to figure out how to word that effectively >_>
Colin Hou
osu! Alumni
4,456 posts
Joined November 2009
Colin Hou
System.ArgumentException: Illegal characters in path.
at System.IO.Path.CheckInvalidPathChars(String path)
at System.IO.Path.GetExtension(String path)
at #vd.#ud.#Pt(Object , #Gc )
at #vd.#ud..ctor(#Gc )
at #1d.#7h.#SW(#Gc )
at #1d.#Oh.#h.#Wi.#s0()
at #Dd.#Ujb.Update()
at #pg.#qh.Update()
at Microsoft.Xna.Framework.Game.DrawFrame()
at Microsoft.Xna.Framework.Game.HostIdle(Object sender, EventArgs e)
at Microsoft.Xna.Framework.GameHost.OnIdle()
at Microsoft.Xna.Framework.WindowsGameHost.ApplicationIdle(Object sender, EventArgs e)
at System.Windows.Forms.Application.ThreadContext.System.Windows.Forms.UnsafeNativeMethods.IMsoComponent.FDoIdle(Int32 grfidlef)
at System.Windows.Forms.Application.ComponentManager.System.Windows.Forms.UnsafeNativeMethods.IMsoComponentManager.FPushMessageLoop(Int32 dwComponentID, Int32 reason, Int32 pvLoopData)
at System.Windows.Forms.Application.ThreadContext.RunMessageLoopInner(Int32 reason, ApplicationContext context)
at System.Windows.Forms.Application.ThreadContext.RunMessageLoop(Int32 reason, ApplicationContext context)
at System.Windows.Forms.Application.Run(Form mainForm)
at Microsoft.Xna.Framework.WindowsGameHost.Run()
at Microsoft.Xna.Framework.Game.Run()
at #pg.#og.#aN()

Have no idea what is this... just playing a map by happy30 :S
Ekaru
3,617 posts
Joined July 2008
Ekaru
Now for some serious input. Some of my Normals are being classified as Hards. So I should probably list them, right? Here's the first one:

http://osu.ppy.sh/s/4258 (Child difficulty)

Why:

Hard criteria: Overall difficulty 6 is greater than 5.
02:11:213 Hard criteria: Note exists 366ms after spinner (less than 400ms).

The OD thing... that was because no one noticed it while modding nor did I notice it. I probably forgot to lower it. ;_; However, all that matters is that this map definitely isn't a Hard.

---------------------

http://osu.ppy.sh/s/5622 (Normal difficulty)

Why:

Hard criteria: Overall difficulty 6 is greater than 5.

This time this was intentional. I think. Maybe. Anyways, it feels right. I just went, "hmm, Normal, let's just go 1 above the 'Normal' OD!" Yeah.

----------------------------------

http://osu.ppy.sh/s/3340 (Normal)

Why:

00:47:416 Hard criteria: Two hitobjects are too far apart (reduce by ~7%).

It's a jump for space and so that you can at least see half of the next slider. Still messy, though.

------------------------------------

http://osu.ppy.sh/s/10331 (Normal)

Why:

01:50:235 Hard criteria: Two hitobjects are too far apart (reduce by ~16%).
00:50:649 Hard criteria: Two hitobjects are too far apart (reduce by ~8%).

Okay, so I used a couple symmetry jumps! ;_; Bah humbug.

---------------------------------

http://osu.ppy.sh/s/7621 (Larto's Normal)

Why:

00:40:370 Hard criteria: A stream exists with 4 objects 175ms apart (less than 180ms).

5 ms? Oh come on. ;_;

--------------------------------

One of my maps has two instances of a circle 375 ms after a spinner. *sniff sniff* This was already covered however.

-------------------------------

http://osu.ppy.sh/s/7201 (Normal)

Why:

00:58:790 Hard criteria: Two objects exist 131ms apart (less than 140ms).
00:58:920 Hard criteria: Two objects exist 130ms apart (less than 140ms).

It's a 115 BPM 1/4 stack of 3 to end the map as opposed to a spinner. It goes to "Di-Gi-Mon", so yeah.

-------------------------------

That's all for now. So yeah. They all had only one or two things, but... :P
Shohei Ohtani
Shohei Ohtani
I wish we could be able to choose what difficulty level AIMod looks for while doing the difficulty thing for a map.

I'm currently getting my insane mode look as a hard. I'd prefer to change thing that help my map become acceptable for an insane mode rather than a hard mode ;~;.

Also what:

Hard: Highest combo length of 13 is greater than 13.

If having a 13 combo is an issue, shouldn't this say "Highest combo length of (#>or=13) is greater than or equal to 13"

~.

Also what the hell why is the highest OD for a hard mode 5? ;~;.
Ekaru
3,617 posts
Joined July 2008
Ekaru
The highest OD for Hard isn't 5, that's the current highest OD for Normal.

"Hard criteria: OD of 6 is higher than 5" means that an OD of over 5 = Hard.

And the "combo length 13 is greater than 13" could be fixed by lowering the 13 shown to a 12, same with the other ones. Problem solved.
Shohei Ohtani
Shohei Ohtani


;~;. I'm guessing it's fine as I intend for this to be an insane diff anyway, but still, for those wanting to make hards~.
Ekaru
3,617 posts
Joined July 2008
Ekaru

CDFA wrote:

(picture)
;~;. I'm guessing it's fine as I intend for this to be an insane diff anyway, but still, for those wanting to make hards~.
...You realize your picture supports what I said, right?

It's a Hard criteria. That means if you have an Overall Difficulty above 5 then it's classified as a Hard.

In fact, your map is classified as a Hard, not an Insane. If it's listing Hard criteria then AI mod thinks it's a Hard. Listing Expert criteria? It thinks it's an Expert. Same with Normal and Insane. If your map is classified as an Easy then AI Mod lists nothing to signify that it thinks it's an Easy.
Shohei Ohtani
Shohei Ohtani
Which is why I want the option to choose which mode AIMod checks it as. It's thinking my map is a hard and telling me things to do to bring it to a hard criteria, when in actuality I want to make it to an insane criteria ~.

Changing the OD to 4 makes the message dissapear, so I'm assuming that that means that any OD above 5 for a map classified as hard is bad, which is kind of low.

I guess I should stop now since you're probably more experienced with this system than I am :3~.
peppy
19,314 posts
Here since the beginning
Topic Starter
peppy

CDFA wrote:

Changing the OD to 4 makes the message dissapear, so I'm assuming that that means that any OD above 5 for a map classified as hard is bad, which is kind of low.

I guess I should stop now since you're probably more experienced with this system than I am :3~.
No, you are thinking about it totally wrong. AiMod is telling you the reasons it is hard. Numbers higher than the limits are not wrong nor out of place. If it was "too hard for hard" it would be judged as insane.

And I doubt there will be support for judging at a certain level, because of the way the system is implemented (it works backwards from the highest difficulty rule-wise). I really think you should be reconsidering the way you create maps if you are looking for what-to-do-to-make-it-insane. The guide should really only need to be used backwards, to dull down difficulty.
peppy
19,314 posts
Here since the beginning
Topic Starter
peppy

MetalMario201 wrote:

I think it would be an improvement to rename the existing difficulty names down one. ie:
Easy → Beginner
Normal → Easy
Hard → Normal
Insane → Hard
Expert → Insane
I would agree with this if it hadn't been three years of using the current difficulty names. I can only see this causing total confusion in the current ecosystem (hell, even thinking about it hurts my brain a bit).
mm201
osu! Alumni
4,495 posts
Joined August 2008
mm201
I don't mean globally, just for the purposes of AIMod difficulty stars. Equivalently, move each of the restrictions down by one difficulty level.
peppy
19,314 posts
Here since the beginning
Topic Starter
peppy
The point is these stars *are* going to be used globally once they are stable.
Jarby
2,963 posts
Joined August 2008
Jarby
I fail to see what is so confusing about changing it to fit MetalMario201's suggestion.
awp
osu! Alumni
11,869 posts
Here since the beginning
awp
I don't really see why it is necessary to scale each difficulty down. Easy is for beginners. On the other hand, there is some ambiguity around which of "Insane" and "Expert" is more difficult. "Insane" is kind of ambiguous in itself, sounding like something that is guaranteed to be over-the-top ridiculous/overmapped or just plain retarded, but given the history it would be a bit jarring to change at this point.
Derekku
COOL DUDE
7,270 posts
Joined March 2009
Derekku

peppy wrote:

I would agree with this if it hadn't been three years of using the current difficulty names.
mm201
osu! Alumni
4,495 posts
Joined August 2008
mm201
What I mean to say is that lowering each of these difficulty margins by one level would put them more in line with community expectations.
Ekaru
3,617 posts
Joined July 2008
Ekaru
Something weird I noticed.

2.5 slider velocity at 85 BPM causes a "Insane criteria: Slider Velocity 2.5 is higher than 2.27".

Yet my 120 BPM map has a "Hard criteria: 2.6 slider velocity is higher than 2.6".

My 165 BPM 2.6 slider velocity map (biggest circles available, this is one of my "let's experiment for the fun of it!" maps) is considered a Hard and doesn't trigger any sort of slider velocity warning.

Is there a reason for this? Because you'd think that the higher the BPM you went the lower the slider velocity would be allowed to be and not the other way around, so I'm confused. ><;
Natteke
6,538 posts
Joined August 2009
Natteke
http://osu.ppy.sh/b/55475

Hyper is harder than Another. WHAT? It's obvious that Another is a lot harder than Hyper, but wtf, it just can't be harder.


What's wrong with that star system... :<
mm201
osu! Alumni
4,495 posts
Joined August 2008
mm201
Personally I don't think combo numbering should be part of the difficulty stars system at all.

Combo numbering can either be right or wrong, but doesn't directly correlate to difficulty. Using them in difficulty calculation will just encourage mappers to use bad combo numbering in order to meet certain difficulty criteria.

If the objective here is to limit the "number of notes in a combo", given that combos should generally be consistent across difficulties, wouldn't limiting the mean number of hit objects/minute be better suited to the job?
RandomJibberish
osu! Alumni
1,833 posts
Joined August 2009
RandomJibberish
On this map AIMod classifies the two harder difficulties which are a Normal and Hardish as Expert and the Easy as a Normal because of spacing issues brought about by the very accurate multiple section timing.

I think this would be fixed if spacing was given a grid or two of leeway. I move beats slightly out of distance snap to align them often my maps, and it doesn't increase the difficulty at all, just makes it look nicer :/
waterfall
28 posts
Joined May 2009
waterfall
I agree that Expert sounds easier than insane. Could you potentially rename expert to "impossible" or something like that? Impossible" certainly sounds harder than insane.

Edit- Maybe lunatic, I dunno. Extreme sounds ok, better to me than expert I guess. As others have said, it seems weird that the other difficulty names are about the difficulty of the map not the skill required to complete it, but "expert" is a description of the player.
Last edited by waterfall , edited 1 time in total.
show more
Please sign in to reply.

New reply

0
1 / 53