Score grade is incorrect when lazer replay is exported and re-imported #24061
Comments
This comment was marked as off-topic.
This comment was marked as off-topic.
No failed plays are involved here. |
This comment was marked as off-topic.
This comment was marked as off-topic.
I did not fail during gameplay (although in this case I only barely managed to pass), even if it seems like it in osu!stable (which isn't osu!lazer).
No, it was played solo. The online score is pretty much the same score as the mentioned replay, except for its lack of replay downloads. |
|
@monochrome22 you're not making any sense at all. @JacksonChen666's score is a lazer score. And a pass at that. What is actually happening here is that the rank is never actually stored to the replay on export. lazer's grade is entirely accuracy-based, but stable's wasn't, and the legacy format decoder uses the legacy logic, causing this to kick in: osu/osu.Game/Scoring/Legacy/LegacyScoreDecoder.cs Lines 225 to 243 in 9b08aaf And in the case above, this would indeed be a D score in stable terms, because accuracy is under 85%. |
bdach
added
area:replay
priority:1
bdach
changed the title
|
Is this fixed? |
|
if you don't know how bug trackers work, the above generally means that it's not fixed |
|
If you look at the screenshot I provided, it looks like it's fixed before one of the members have marked it as fixed. |
|
Please clarify your intention next time, and avoid posting screenshots without making sure it shows what you're referring to. That being said, the accuracy cutoffs has indeed been adjusted in #25601 to match the code snippet mentioned in #24061 (comment). However, this has no relevance in the issue thread's main point. As per the title of this thread, the problem is that the score rank gets populated with stable's logic, rather than lazer/ It seems possible to not have stable's logic run for lazer scores using the diff --git a/osu.Game/Scoring/Legacy/LegacyScoreDecoder.cs b/osu.Game/Scoring/Legacy/LegacyScoreDecoder.cs
index ed11691674..4d651b2fb2 100644
--- a/osu.Game/Scoring/Legacy/LegacyScoreDecoder.cs
+++ b/osu.Game/Scoring/Legacy/LegacyScoreDecoder.cs
@@ -132,6 +132,9 @@ public Score Parse(Stream stream)
PopulateAccuracy(score.ScoreInfo);
+ if (!score.ScoreInfo.IsLegacyScore)
+ score.ScoreInfo.Rank = currentRuleset.CreateScoreProcessor().RankFromAccuracy(score.ScoreInfo.Accuracy);
+
// before returning for database import, we must restore the database-sourced BeatmapInfo.
// if not, the clone operation in GetPlayableBeatmap will cause a dereference and subsequent database exception.
score.ScoreInfo.BeatmapInfo = workingBeatmap.BeatmapInfo;But upon testing that in real leaderboards, it appears that the accuracy is different on import, potentially being high enough to display a different rank instead: 
(left is online, right is post-import) |
Did you stop investigating at this point? It's a weird place to leave things, just means the next person to follow up is going to have to repeat and continue where you left. |
peppy
added
priority:0
|
Bumping to priority 0. Having broken replay round trip is going to cause a lot of headaches as users start to move to lazer. |
Closes ppy#24061. The gist of this change is that if the `LegacyReplaySoloScoreInfo` bolt-on is present in the replay, then it can (and is) used to recompute the accuracy, and rank is computed based on that. This was the missing part of ppy#24061 (comment). The accuracy would change on import before that because the encode process is _lossy_ if the `LegacyReplaySoloScoreInfo` bolt-on is not used, as the legacy format only has 6 fields for encoding judgement counts, and some judgements that affect accuracy in lazer do not fit into that. Note that this _only_ fixes _relatively_ new lazer scores looking wrong after reimport. - Very old lazer scores, i.e. ones that don't have the `LegacyReplaySoloScoreInfo` bolt-on, obviously can't use it to repopulate. There's really not much good that can be done there, so the stable pathways are used as a fallback that always works. - For stable replays, `ScoreImporter` recalculates the accuracy of the score _again_ in https://github.com/ppy/osu/blob/15a5fd7e4c8e8e3c38382cfd3d5d676d107d7908/osu.Game/Scoring/ScoreImporter.cs#L106-L110 as `StandardisedScoreMigrationTools.UpdateFromLegacy()` recomputes _both_ total score and accuracy. This makes a _semblance_ of sense as it attempts to make the accuracy of stable and lazer replays comparable. In most cases it also won't matter, as the only ruleset where accuracy changed between the legacy implementation and current lazer accuracy is mania. But it is also an inaccurate process (as, again, some of the required data is not in the replay, namely judgement counts of ticks and so on). For whatever's worth, a similar thing happens server-side in https://github.com/ppy/osu-queue-score-statistics/blob/106c2948dbe695efcad5972d32cd46f4b36005cc/osu.Server.Queues.ScoreStatisticsProcessor/Commands/Queue/BatchInserter.cs#L319 - However, _ranks_ of stable scores will still use the local stable reimplementation of ranks, i.e. a 1-miss stable score in osu! ruleset will be an A rather than an S. See importer: https://github.com/ppy/osu-queue-score-statistics/blob/106c2948dbe695efcad5972d32cd46f4b36005cc/osu.Server.Queues.ScoreStatisticsProcessor/Commands/Queue/BatchInserter.cs#L237 (it's the same method which is renamed to `PopulateLegacyAccuracyAndRank()` in this commit). That is all a bit of a mess honestly, but I'm not sure where to even begin there...
Closes ppy#24061. The gist of this change is that if the `LegacyReplaySoloScoreInfo` bolt-on is present in the replay, then it can (and is) used to recompute the accuracy, and rank is computed based on that. This was the missing part of ppy#24061 (comment). The accuracy would change on import before that because the encode process is _lossy_ if the `LegacyReplaySoloScoreInfo` bolt-on is not used, as the legacy format only has 6 fields for encoding judgement counts, and some judgements that affect accuracy in lazer do not fit into that. Note that this _only_ fixes _relatively_ new lazer scores looking wrong after reimport. - Very old lazer scores, i.e. ones that don't have the `LegacyReplaySoloScoreInfo` bolt-on, obviously can't use it to repopulate. There's really not much good that can be done there, so the stable pathways are used as a fallback that always works. - For stable replays, `ScoreImporter` recalculates the accuracy of the score _again_ in https://github.com/ppy/osu/blob/15a5fd7e4c8e8e3c38382cfd3d5d676d107d7908/osu.Game/Scoring/ScoreImporter.cs#L106-L110 as `StandardisedScoreMigrationTools.UpdateFromLegacy()` recomputes _both_ total score and accuracy. This makes a _semblance_ of sense as it attempts to make the accuracy of stable and lazer replays comparable. In most cases it also won't matter, as the only ruleset where accuracy changed between the legacy implementation and current lazer accuracy is mania. But it is also an inaccurate process (as, again, some of the required data is not in the replay, namely judgement counts of ticks and so on). For whatever's worth, a similar thing happens server-side in https://github.com/ppy/osu-queue-score-statistics/blob/106c2948dbe695efcad5972d32cd46f4b36005cc/osu.Server.Queues.ScoreStatisticsProcessor/Commands/Queue/BatchInserter.cs#L319 - However, _ranks_ of stable scores will still use the local stable reimplementation of ranks, i.e. a 1-miss stable score in osu! ruleset will be an A rather than an S. See importer: https://github.com/ppy/osu-queue-score-statistics/blob/106c2948dbe695efcad5972d32cd46f4b36005cc/osu.Server.Queues.ScoreStatisticsProcessor/Commands/Queue/BatchInserter.cs#L237 (it's the same method which is renamed to `PopulateLegacyAccuracyAndRank()` in this commit). That is all a bit of a mess honestly, but I'm not sure where to even begin there...
Type
Cosmetic
Bug description
I imported a score of my own, clicked to view then saw a contradiction. (See screenshot)
I don't know how to better describe these things in the title.Map
Replay (in a .zip)
Screenshots or videos
Version
2023.621.0-lazer
Logs
database.log
input.log
network.log
performance-audio.log
performance-draw.log
performance-input.log
performance-update.log
performance.log
runtime.log
The text was updated successfully, but these errors were encountered: