18 Dec 2023
osu!(lazer) Updates & Preparing For Ranked Play
by Walavouchey

osu!(lazer) Updates & Preparing For Ranked Play

by Walavouchey

For this last update of the year, the transition to ranked play in the new client is about to begin its final ascent.

They say learning through multiple media is optimal. While these short text summaries of the changes in each update serve in large part just to supplement the main video, this time it may be helpful to have detailed explanations both in text and video format. For optimal understanding, you may want to both watch the video and use the summary below as reference.

The gist is that the team is looking for more people to try out osu!(lazer) before game mechanics are locked and performance point leaderboards are turned on, but before that, let's take a look at the new changes:

Smaller changes

  • Dropdown menus are now searchable. This mainly means that skins are now searchable, but it also applies to every dropdown menu in the game.
  • Mod selection search box is now auto-focused by default. The new paradigm of quickly selecting mods is to hit F1, (partly) type its name, then pressing Enter. If you prefer the old way of selecting mods, uncheck Automatically focus search text box in mod select in the settings.
  • Use these mods option is now visible when clicking anywhere on a leaderboard score panel. This neat feature has existed for a while now, but the problem was that nobody knew about it. Now you know.
  • Song select search performance optimisations. It should be around 90% faster or so peppy says.
  • Multiplayer hosts can now abort matches. You can find the big red abort button in the lobby. (No !mp command support yet.)
  • Mod multiplier adjustments. Half Time changed from 0.7x to 0.3x, Classic changed from 1x to 0.5x, and Synesthesia changed from 1x to 0.8x. These are low "placeholder" values to prevent these mods that make the game easier from ranking high on beatmap leaderboards, pending further balancing.
  • HP drain changes. Following last release's change to the osu! game mode, osu!mania HP drain now matches osu!(stable). In addition, the osu! mode's HP drain is reverted to the previous osu!(lazer) algorithm by default (with some improvements), while the osu!(stable) algorithm is still present with Classic mod enabled.

The full changelog can of course be found where it always resides, but for this time, let's highlight in a little more detail a few changes that are either exciting or that you should probably know about.

Optimised freehand slider drawing

The last release introduced freehand slider drawing, both to the excitement and horror of mappers. The initial implementation had lots of jank to be ironed out, but with this release it works pretty alright or at least better than before:

Scoring changes

Combo is nerfed (for high combo). That's the short explanation.

To elaborate, in osu! with ScoreV1 and ScoreV2, the cost of a miss grew fast for higher and higher combos (quadratically, i.e. combo^2). This implied that a single miss could completely ruin the score of a play and cost up to 350k points in ScoreV2.

In order to reduce this effect, a formula change was applied to make the cost of a miss grow at a slower pace (combo^1.5), which reduces the miss penalty at high combos while keeping it high at low combos, in a similar spirit as how combo is capped in osu!taiko or osu!catch but without any hard cap.

This change is introduced together with a complete rebalance of the combo–accuracy split based on user feedback, further making accuracy a bigger part of the score. Note that the new formula will imply higher scores overall and might require a bit of time getting used to.

Compared with ScoreV2 (the previous system in osu!(lazer)), with this change, high-accuracy plays with low combo can be expected to see a score increase upwards of around 200k–300k points. For a detailed comparison between ScoreV1, ScoreV2 and the new scoring system, see this spreadsheet, or maybe look at the actual code change if you understand better that way.

Adjusted slider mechanics

Two changes have been made to make sliders more lenient.

When clicking a slider early, the follow circle will now immediately start in a tracking state even if the cursor leaves the slider ball before the slider starts. It's easiest explained in video form:

Secondly, for very fast reverse sliders, clicking late would mean missing on the ticks or repeats, so players would click early to compensate. However, sliders in osu!(lazer) judge the accuracy of the initial hit, inevitably causing low accuracy on these types of sliders.

With this change, there is no longer a need to click early to avoid a miss, as ticks or repeats will be completed even for late hits.

Grade criteria adjustments for osu!catch

In osu!(stable), the accuracy (and judgement) requirements for each grade are as follows:

Grade osu! / osu!taiko osu!catch osu!mania
SS 100% 100% 100%
S ≥90% (≤1% MEHs/50s, no misses) ≥98% ≥95%
A ≥80% (no misses) or ≥90% ≥94% ≥90%
B ≥70% (no misses) or ≥80% ≥90% ≥80%
C ≥60% ≥85% ≥70%

Meanwhile, osu!(lazer) has these accuracy cutoffs now:

Grade osu! / osu!taiko / osu!mania osu!catch
SS 100% 100%
S ≥95% ≥98%
A ≥90% ≥94%
B ≥80% ≥90%
C ≥70% ≥85%

In this update, osu!catch has been changed to match osu!(stable) due to it being easier to get higher accuracies there.

Moreover, these current grade criteria will be locked, meaning they're not intended to change for the foreseeable future. While the no-miss requirement for S in the osu! game mode has made it easy to tell full combos apart (ignoring slider breaks), the plan is to prominently display full combos in a different way.


The important part here is that your feedback on these mechanics will be especially helpful during the next two weeks as the developers get ready to enable performance points for the new client. After trying it out here to see how it currently feels, You can join the discourse in this forum thread or this GitHub discussion.

Feel free to ask any questions about this update or the direction of osu!(lazer) in general in the comments of this post.

—Walavouchey

Comments105

Sort by

W

i will switch to lazer once lazer's input latency feels the same or better than stable
i've been switching between the two and i can confidently say lazer has a bit worse input latency
noticeable enough to the point where my brain has to compensate for it and it's mentally draining

hopefully upcoming optimizations would have a noticeable impact on my pc, nevertheless lazer's looking great so far

Fence_2 Commanvil

I have a pretty weak CPU, but my latency in stable is 0.35 ms. However in lazer it's 1 ms. I feel like the game is a little bit off, but not too much. After 1 day of playing on lazer, I got completely used to playing on lazer. Sometimes the game freezes, but I was able to set a new top 2 score for myself. So it's pretty playable..

My tip is to try playing around with different frame limiters.
For me the best one is 4x refresh rate. Even though I play on a 165 hz monitor and I hate frame limiters :D

IsbjabadnamE Commanvil

yeah if lazer is better optimized than stable I will switch too

Spektacle IsbjabadnamE

From my experience playing from Stable on Windows and Lazer natively on Mac, it feels the same, if not a little bit better. I've been using Lazer as my main for a while now and don't regret it one bit.

yippee hooray! the world has been saved!

One more thing. Do combo nerf affects pp calculation ?
As I understand this affected by accuracy calculation. and combo considered more valuable than perfect timing (and this is wierd because perfect timing is harder to perform than combo hold). Especially on long maps (really do not understand why short map 1 minute length with full combo gives more pp than big 3-5 minutes and harder map with 1-2 misses in the middle).
Perhaps make combo loss less punishing, at least on big combo (let's say pp loss when breaking 100+ combo should be equal to pp loss at breaking 1000+).
My opinion combo length should now be counted for pp accuracy penalty at all, or at least it should be weighted a lot less than hit accuracy (which is a lot harder).

Playing osu lazer on linux from time to time. Well, it works. But...
I do not know why but in stable (OSU stable + wine + pipewire/alsa) I got higher accuracy score than in osu lazer (same skin same map). Feels like Lazer sometimes register keyboard input with offset or draw circle with offset, just some circles get 100 score instead of 300. Input in laser have bigger dispersion than stable.
And whis is not a system falture, tried this with different distros with different driver versions on pulseaudio and pipewire.

Grey4140 Yeriko

The AppImage doesn't work for me and you too? Well I will try with wine.

-infinite- Yeriko

slider heads count for accuracy in lazer (unless using classic mod), so even with same UR and 0 offset one may get lower acc in lazer.

Please put a banner in-game in Stable making it very clear you need feedback about permanent changes to the game. A lot of people will not otherwise understand or just don't look at the website. This is too important to be so relatively hidden.

Walavouchey Happy Satoko

yes the devs want to finalise game mechanics and move on to other areas of the game, but even after pp is enabled, any required changes will be made. things will likely be unstable for a while, scores may be deleted etc. please watch the last part of the video again.
devs are getting plenty feedback already (and can't really keep up with everything), and there will be even more of that as people switch over to lazer.
rest assured there will be a news post, @osugame tweet, youtube video (except likely not for the end of the year as they're taking a break, as stated the video) to cover this, and likely content creators as well. anyone who would care will know.

Happy Satoko Walavouchey

alright, thanks for the assurance

tried for the first time, am not going back to stable

que legal

when does pp start counting?

maybe you should watch the video again :)

-Kori txvii

whenever peppy announces lazer is ready

Please optimize Osu!Lazer for pc with intel celeron too like how Osu! is working in my pc (intel celeron n4500 and integrated graphics) right now

can't wait for the Ranked🔥🔥🔥

hats off to the entire dev team. I love the way you're moving osu forward, please keep this up!

After this update, so many little red crosses appeared when I play. Is there someone who know how I can remove them ?

ierapir Dilyann

You can still play with Argon pro skin

Sleepteiner Dilyann

Those little red crosses indicate that the player has missed a slider tick or a slider end. It is very important feedback for the player.

However, the current design of it is a placeholder and there will probably be more customization of it available in the future.

ahhhh shiet lazer editor lookin' really good

xbox dogdeleted
Walavouchey xbox dog

it has been changed to 0.96x

real

orangix Windows Me

windows is so fake for this comment frfr

Windows Me orangix

real

Aren't they gonna add rank and pp at the bottom like stable 🧐❓

Will be possibile to transfer stable skin 1:1 to lazer in mania?

have you tried it?

M1si3k Walavouchey

yes and it sucks

can we have an option for osu lazer to look like osu stable? lazer runs so bad for my pc for some reason. one more question, can we have ranked score v1 option (ranked) with score v2 (ranked) too?

There are too many new mechanisms in Lazer now. I feel like that it needs much more time to get used to it if you spend too much time on stable. It may hurt.

i like everything about lazer except for the fact that it runs like ass

Daoren advu_

lazer runs better than standard for me

I have the perfect disaster solution for this problem and its been done so many times before anyway. It might sound radical but split the community in two and create a new webpage for Lazer users and let them have their own server.

In memory of Q1/QW, SC/BW, D2/LOD. I Remember Civ 4 even doing it over and over like four times with various expansions.

Happy Yule&a good new year!

orangix l1mi

lazer.ppy.sh

I am very excitement!

HISTORY IS ABOUT TO BE MADE

olibomby is my hero

Nobody blames this because of the bugs, not because of the points system or the nerfs or buffs that it entails

we are un-gatekeeping nice sliders

is osu!lazer gonna have a pp rework? (since after checking/rewatching my top play on lazer, it was worth 8pp more. Multiple of my plays have different pp on lazer, than on stable)

Nitobliss vanikou

That's because acc can be dropped on slider heads now

vanikou Nitobliss

Ooooh, that makes sense. Thank you ^^

what a time to be alive

So uhhh, first is pp and then next year some optimization?

while some areas haven't been optimised a whole lot, like storyboards or some operations in the editor, performance optimisation is on-going and happens pretty much every release. if you see issues you should report here along with hardware/system info, log files and which game version you're using

Are there any plans about osu! Lazer supporting songs backup in nearly future?

Walavouchey dokuro_piko

if i understand your question correctly this has been asked about repeatedly (e.g. here and here) but the answer has been no. you should back up your installation folder by external means

dokuro_piko Walavouchey

According to the second link it answers my question, and it is yes: "Eventually the game will support that yes. If any files are missing you will have the option to repair the beatmap". Thanks.

alyux dokuro_piko

I wanna have that too

I have a few comments that are unrelated to each other and don't quite fit into the feedback thread, so I'll post them here.

First, two weeks is far too short to have a sufficient community discussion about all of these issues. Many people with important perspectives will not have time to completely get those perspectives across in that time.

Second, the change to the score system is a significant improvement over scorev2. I still think the score system is bad since I dislike any combo influence at all, but an improvement is an improvement.

In order to make it clear how to maybe improve the score system more, I very much support the survey idea mentioned in the video. And, either with the survey or some time before it, there needs to be a clearer and more accessible representation of what the new scoring system actually changes in practice. The spreadsheet is great, but the majority of people will likely have a difficult time understanding what any of those numbers really mean in practice. I think the best way to do this is to provide comparisons of what the scoring changes will do to leaderboards with a ton of variety of miss count and accuracy on the scores on the leaderboard so people can see the difference of how different kinds of scores are valued.

A couple types of good leaderboards to use are ones of popular extremely high star rating maps and also niche extremely hard lower star maps. Here are some potentially good ones:

https://osu.ppy.sh/beatmapsets/396221#osu/862088
https://osu.ppy.sh/beatmapsets/295422#osu/663725
https://osu.ppy.sh/beatmapsets/1102672#osu/2303523
https://osu.ppy.sh/beatmapsets/1843396#osu/3786497 (This one might actually have too few scores to be valuable here)

Third, the part of the video that conveys that "some people dislike slider accuracy so we made these two slider changes" feels quite disingenuous considering that those two slider changes, while very good, are really just addressing edge cases and that the core problem remains pretty unchanged. The core problem being that there are thousands of ranked maps using slider accuracy leniency as an intentional mapping tool, so forcing slider accuracy on those maps as the new default clearly hurts those maps and makes people want to play them less.

This being a real problem is proven by how many people were complaining in the playlist chats about Lana Del Rey - Cinnamon Girl [Don't Hurt Me] in the recent playlist B of the Autumn 2023 Beatmap Spotlights being not fun to play due to being hard to acc. The majority of the hardest bpm and rhythm changes in that map probably (but I could be wrong for this specific map) intentionally used sliders to make the rhythmic reading difficulty spikes less unfair. Seeing people blame the mapper of that good map for problems with the map that are not the mapper's fault at all in a project that is supposedly about promoting the appreciation of high quality maps is pretty depressing.

How the problem of slider accuracy is phrased in the video feels similar to the tactic used to dismiss people who are against slider accuracy being forced on maps made in stable as the new default which is to claim that being against slider accuracy is to stand in the way of "progress." Forcing slider accuracy on maps made in stable is not "progress" in any objective sense. It is a subjective change decided to be put into the game by a relatively small group of people who are obviously well meaning, but have a significant history of not understanding the full implications of gameplay changes that they intend to put into the game. If this is the direction for the game that the player and mapper communities really want, then so be it, but to phase the real concerns with slider accuracy as a relatively trivial thing that people will just get used to feels disrespectful.

Happy Satoko Sleepteiner

Really hope they listen to you. Whatever happens, thank you so much for speaking up. I agree 100%

l1mi Sleepteiner

ScoreV2 have a fundamenally critical flaw.. its boring.

Adryzz l1mi

yes, which is why you can use classic scoring in lazer

orangix l1mi
Walavouchey Sleepteiner

First, two weeks is far too short to have a sufficient community discussion about all of these issues. Many people with important perspectives will not have time to completely get those perspectives across in that time.

read this discord message by peppy (osu!dev server)

to add to that, the idea is that feedback from the most amount and widest spectrum of players possible will be the most valuable, which is most effectively achieved simply by having lots of people try lazer

In order to make it clear how to maybe improve the score system more, I very much support the survey idea mentioned in the video. And, either with the survey or some time before it, there needs to be a clearer and more accessible representation of what the new scoring system actually changes in practice. The spreadsheet is great, but the majority of people will likely have a difficult time understanding what any of those numbers really mean in practice. I think the best way to do this is to provide comparisons of what the scoring changes will do to leaderboards with a ton of variety of miss count and accuracy on the scores on the leaderboard so people can see the difference of how different kinds of scores are valued.

comprehensive comparisons would be very hard to do before leaderboards are combined, but once that happens in a few weeks it'll be trivial to tell how scores are affected since the leaderboards will simply look different. you can then compare in-game on stable (where nothing should change afaik) or with screenshots/wayback machine

also we'll see how mod multipliers will affect this (namely classic mod being 0.5x but it looks like that'll be bumped to 0.96x), but those can be changed on the fly later

[...] those two slider changes, while very good, are really just addressing edge cases and that the core problem remains pretty unchanged. The core problem being that there are thousands of ranked maps using slider accuracy leniency as an intentional mapping tool, so forcing slider accuracy on those maps as the new default clearly hurts those maps and makes people want to play them less.
[...]
Forcing slider accuracy on maps made in stable is not "progress" in any objective sense. It is a subjective change decided to be put into the game by a relatively small group of people who are obviously well meaning, but have a significant history of not understanding the full implications of gameplay changes that they intend to put into the game. If this is the direction for the game that the player and mapper communities really want, then so be it, but to phase the real concerns with slider accuracy as a relatively trivial thing that people will just get used to feels disrespectful.

well yeah if you disagree with slider accuracy on a fundamental level then that's that. some people are fine with it, some aren't, and some don't really care, but we'll see how many people actually complain about this when more and more people decide to try moving over to lazer

and in general see this answer by peppy on a related proposal. classic mod is a thing and stable will be continue to be supported until almost no one plays it anymore, for this exact reason

Sleepteiner Walavouchey

read this discord message by peppy

I don't see how this changes much. It seems like he is saying "sure we MIGHT change something after these two weeks, but we might not, so if you don't give feedback now, you might lose your chance to have an impact." This is what I'm against since two weeks is not enough time for everyone who might want to give their full feedback to do so. But, this seems set in stone, so not much I can do, I guess.

once that happens in a few weeks it'll be trivial to tell how scores are affected

Right... But that is after the two week window where good feedback is apparently very important. I don't see how we are going to get that much good feedback on the scoring systems in the next two weeks without a way for players to engage with the changes in a more practical way...

and in general see this answer by peppy on a related proposal. classic mod is a thing and stable will be continue to be supported until almost no one plays it anymore, for this exact reason

Just like clayton said in response to that comment, peppy's comment doesn't actually address any of these issues. Stable still existing and classic mod being a thing is not and will never be the answer to these problems.

In the specific case of the problems I'm bringing up, this is because the vast majority of players will not be using the classic mod which means that the vast majority of players will enjoy maps using certain mapping concepts much less than before. If players enjoy certain mapping concepts less, then mappers will be pressured to use those mapping concepts less. This will force mapping itself to be narrower in variety which will leave a permanent scar on the game's history and as a result, limit the game's potential.

This is all pretty esoteric, so most players will not think about any of this. Which means that it is really up to the dev team to decide how much they want to care about leaving scars on the game's history compared to their idea of what progress is.

Also, I don't think forcing classic mod on certain maps is a good idea either since most players want to be able to customize the maps as much as possible. If that customization is taken away, people will avoid those maps, creating a very similar problem to the problem that forcing slider acc on them by default creates.

Wispy Sleepteiner

PREACH

May I ask to the team that, any plans to add combo burst as an optional option like osu! stable?

Other than that, I feel the accuracy system doesn't affect me much (osu! catch), but the scoring combo nerf is a good change! It'll hopefully won't make players end up getting <500k w/ only 1 - 2 misses. Let's hope the balance goes well tho

Walavouchey Alpha Green Mg

adding combo bursts is tracked here https://github.com/ppy/osu/issues/4629
someone has written an implementation here https://github.com/ppy/osu/pull/17616 but it hasn't made it past review yet

Alpha Green Mg Walavouchey

Oh yeah, the issue #4629, the one I'm following, thanks for the remind!

if I now farm pp in osu lazer, will this carry over to release?

Walavouchey TheMov1eOne

the answer is maybe. i'll quote a discord message from peppy (from the osu!dev server https://discord.com/channels/188630481301012481/188630652340404224/1185083869037330502)

Consider the first few months after we switch things on to be somewhat of a testing ground. There might be a need to delete some scores. Of course, we’ll try our best not to, but if there is significant pushback from the larger community on game mechanics which can only be recovered by removing scores, we will prioritise getting-things-right.

Miyuki1deleted
Walavouchey Miyuki1

you're talking about song select and the results screen, right? it's planned to be a thing eventually, but will most likely use the new skinning system first and foremost (i.e. the skin editor as it gets fleshed out)

did audio get fixed

Walavouchey fieryrage

it's fine on my end. i get lower audio latency on lazer than on stable. if you encounter problems please report here along with hardware/system info

I am so excited for this, but also dread the thought of lazer being locked from getting core gameplay updates :/

at some point yes, as a "for the foreseeable future" thing, but if things break and if people complain, fixes and changes to game mechanics as well as score wipes after pp is enabled can happen (and most likely will or so i'm told). time will tell though - we're treading unknown waters.

on the bright side, not having to worry about gameplay, which has been the focus during this year, will enable the devs to move on to other areas (such as beatmap submission, skin editor, tournament support, the oh-so-many feature requests, etc. etc.)

just want to leave a step before things changes. early happy christmas everyone! o/

Hate that difficulty adjust and bpm adjust are gonna be ranked. Absolutely no incentive to learn control or higher/lower ARs is gonna be so great for new players.

fleuphy PepperJack

even on stable there is no incentive for non-tournament players to learn them if they are just trying to gain rank. People will still have to learn it if they want to do well in tournaments and people will still find interest in those skillsets, but they never have and probably never will be favored in the performance algorithm (with the exception of AR11 for DTHR plays)

I for one think opening up diff adjust and bpm adjust to rankings will allow for some crazy things. Players like Aery will actually be able to submit their crazy negative AR scores to the leaderboard since there is official support for their playstyle. Players with extreme memory or reaction time can do 2x speed on maps, breaking the AR11 barrier in the process and allowing for even crazier plays.

bostick PepperJack

Forcing people to be limited by what AR and BPM they prefer to play is worse for the game. ALL play styles should be viable and fleshed out for the best player experience. Osu is a preference game so why on earth do players have to deal with uncomfortable AR or density on their favorite songs or a map they like. This is a really good addition

Duck o-o PepperJack

Theres no incentive anyway
Ive seen so many people just ignore good low ar maps because "thats too low" when they are like 5 digit and its a 4 star lul

shoutouts to olibomby making osu mania in scratch love that dude fr

olibomby is like king midas to whatever he touches it's insane

can you believe it guys? christmas just a week away christmas is in a week woohoo. i am so happy about this information christmas just a week away oh wow can you believe it? christmas just in a week it got here so fast christmas just a-

Awesome

I started playing osu! on lazer, and then I switched to stable a few months later. I have to say it feels weird now that lazer will be ranked soon.

I think the main thing still preventing me from using lazer is the filesystem hierarchy, I have duplicates of some maps which are different from one another and have different replays (some of which I hold dear). (I think that the migration dedupes beatmap IDs, and that just doesn't work for me. I know I can change the beatmap ID, then edit the replay file and change the md5, but that's a lot of steps) Also harder to do GDs, hitsounds, etc., and those seem pretty far off as client features on lazer.

So that's my current problems with lazer...

Also, lazer crashes my GNOME session, if someone else is experiencing this can u bug report thx

tromino orangix

haven't had any crashes so far... though since gnome 45 I've noticed the game visuals going off-screen whenever I open the activities overview (though just pressing alt+tab while in-game fixes it)

- Marco - orangix

You can send a bug report here

orangix - Marco -

yeah im lazy ;p

Freehand slider drawing looks even more amazing now.

peppy is an awesome developer and should be recognized for his efforts listening to the community and working in the game development so heavily and greatly.

Skubi Mekki

Thank OliBomby for the freehand slider implementation

the implementation was done by Tom94 and default0, while the optimisations in this update were by OliBomby

orangix Mekki

To be honest when I heard about this I was horrified because I was imagining those sliders with a million control points but it's really smooth now I actually see it

HYYYYYYPEEEEEEEEEE

/

HYYYYYYPEEEEEEEEEE