Skip to content

This Time For Real - #31

Merged
TextAdventurer12 merged 15 commits into
TextAdventurer12:fc-time-limitfrom
kwotaq:fc-time-limit
Apr 17, 2026
Merged

This Time For Real#31
TextAdventurer12 merged 15 commits into
TextAdventurer12:fc-time-limitfrom
kwotaq:fc-time-limit

Conversation

@kwotaq

@kwotaq kwotaq commented Apr 17, 2026

Copy link
Copy Markdown

the pop deev

stanriders and others added 15 commits March 24, 2026 17:57
Now that agility isn't "speedaim" that covers both snap and flow we can
restore it back into it's original scaling. Arguably it should be _more_
than d/t^2 but that's a thing to explore separately
At this point maybe having a shared small circle bonus in ODHO isn't
even worth it since we have different d/t scalings in all evaluators.
Bonus was made for snap (~d/t^1.65), here it's adjusted to be higher on
agility (~d/t^2) and lower on flow (~d/t). Practically affects like [1
map](https://osu.ppy.sh/beatmapsets/2191876#osu/5192354)
This buffs longer jump sections and slightly nerfs short jump spikes.
Mostly a way to slightly restore non-aimslop DT aim maps since they got
hit with all the nerfs a bit too hard

---------

Co-authored-by: James Wilson <tsunyoku@gmail.com>
This rework target is C-type nerf, as this map gets absurd amount of
reading pp, because angles are considered very unrepetitive. What
results in this score being worth absurd 1.4k pp.

This PR is up to heavy discussion because it can be made much more
general, touching more maps. So I wait on pp committee opinion on what
of the parts can be removed.

Current checks for angle to be nerfed:
- The smaller angle has to be very sharp: <20 degrees, full power on <5
degrees
- The larger angle has to be wide: >60 degrees, full power on >120
degrees

If pattern meets all the criteria - it would be considered repetitive.
For now practically no map meets this criteria to significant amount
except C-type.
- Closes ppy#25254

Solves chunking-related issues by starting a new chunk for every object,
and allowing chunks to be different lengths. This is retrofitted to
existing calculations, and doesn't have other ramifications in the way
something like tr3s's continuous strains do.

The effect of this can be seen in the video below. In live pp, the map
would lose sr when increasing the rate at certain points, but in the
video, the sr rises consistently as expected.


https://github.com/user-attachments/assets/2d58946a-9e0c-4f6b-912a-71dfe75c0f8a

A multiplier has been added inside `DifficultyValue()` to account for SR
being slightly lower from more granular summation.

When testing this rework in PerformanceCalculatorGUI, I recommend using
[this
fork](https://github.com/Finadoggie/osu-tools/tree/variable-length-strains).
I can't guarantee that the visuals it shows are correct, but they are
definitely more correct than not.

Edit: Use [this
fork](https://github.com/Finadoggie/osu-tools/tree/unsynced-variable-length-strains)
for testing now. Old one is now for testing
SynchronizedVariableLengthStrainSkill

---------

Co-authored-by: tsunyoku <tsunyoku@gmail.com>
Co-authored-by: StanR <8269193+stanriders@users.noreply.github.com>
This doesn't affect values as per the original refactor.

Co-authored-by: StanR <8269193+stanriders@users.noreply.github.com>
…calculation (ppy#37184)

Adjusting difficulty before we calculate the flow probability is
technically wrong. This makes flow adjustment on RX actually work as it
should instead of trying to lower the snap to achieve similar results.
TD flow buff was removed because here it actually made some maps _very_
buffed. Values are mostly the same otherwise
This PR moves the addition of the bonuses closer to their declaration
and also renames some variables and functions to more closely resemble
the way they're used. The names can be discussed for I'm not sold on
them either.

---------

Co-authored-by: James Wilson <tsunyoku@gmail.com>
Currently all low spaced (<radius) aim is evaluated as low difficulty
snap. After we separated snap, flow and agility we don't actually need
snap aim to be reduced for low distance - this is something that flow
aim should cover instead as low spacing is guaranteed to be either
flowed or not aimed at all. By extension this fixes the need for the
flow velocity bonus to be extremely high as it no longer clashes with
snap when the changes happen between low and high spacing.
# Conflicts:
#	osu.Game.Rulesets.Osu/Difficulty/Skills/Aim.cs
@TextAdventurer12
TextAdventurer12 merged commit 24dbc71 into TextAdventurer12:fc-time-limit Apr 17, 2026
4 of 14 checks passed
@kwotaq
kwotaq deleted the fc-time-limit branch April 17, 2026 12:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants