Make mania award SS even if there are GREAT judgements #29640
Conversation
This should cover it no? https://github.com/ppy/osu-queue-score-statistics/blob/9482429197400375d0a9923dfca4454f62aaec85/osu.Server.Queues.ScoreStatisticsProcessor/Commands/Maintenance/VerifyImportedScoresCommand.cs#L241-L246 |
|
That command only applies to imported scores right? This needs to be applied to lazer scores also. |
|
Right, okay, that is true. |
|
So acc can still be below 100 and have SS? Why not also make 300s give 100 acc? It would still fix SS from requiring a million points, but it would also keep old scores from having their acc/grades nerfed. |
|
Because greats already give the maximum score in the combo portion, so doing the same in accuracy will mean a score with 100% greats will be 1M score. The next question is "why don't you move the 305 score to the combo portion", and the answer there is because getting greats will be more punishing given that that portion is scaled by the current combo. Also see the many discussions linked where the above was decided: #25945 |
|
I was talking about the displayed accuracy, not the portion of points given. MAX should still give more points and all 300 shouldn't give 1m. I meant that the accuracy percentage should work how it does in stable so SS scores stay 100% and other scores don't get lowered. Unless that's how it already works here and I misunderstood the code, in which case I apologize. |
|
Yeah no. That's simply not how the game works. I will not be adding yet another hidden value to scoring. Accuracy is calculated based on the scoring portion, so the only acceptable change would be to move 305 to the combo portion. |
|
I would rather that than people having an SS with 98%. |
|
I'm not making a call on that. It will need a community discussion. |
|
I had already said that in discussion for more than 3 times or so. In osu!mania if wanna get All rainbow 300 SS, for SS players, its actual OD is much higher than 10 under OD 8 with lazer scoring, not really fair for mania players isn’t it. I sent that in discussion but not a PR simply because I programs mainly in C, not C#, or I would have made a PR. Who said yellow 300 is not 300? That difference should only be seen while calculating PP. I also had already said that SS without 100 accuracy is also acceptable. But at last, peppy voted on the wrong place — twitter, without ringing anyone, especially someplace where people are banned from using twitter, China, but it has a very large amount of active mania players; without saying that great is actually yellow 300, while that is referred to 200(green) actually in community. Many people were mislead, another issue may be solved from voting that change on osu.ppy.sh and make a banner ad at the bottom of the first screen both osu!lazer and osu!stable. At the end, I really favor the change. I had seen that many players laughed at the change in osu!mania that SS players whose ranked plays are all SS, but resulting a 99% accuracy, including the player being launched at took that issue seriously, think that ironic. The thing that humans are not machines, SS can be lost at a very very slight deviation, has to be reminded. Many beatmaps with higher star rating, just say 5*, has no 1,000,000 from the time they were ranked. |
|
@Adrriii At this point I'm going to rope you in, in case you have any thoughts. I believe this PR is still in line with original discussions... If the above suggestions is the path instead then it's looking like a post mortem on how not to do scoring, because we can't easily/exactly recalc those scores. |
|
I'll just say that there was a poll on the forums, not on twitter, and people seem fine with fixing just the rank part of it there while keeping the accuracy part as is. Yes the n=79 sample size is small, it's still more than n=2 of passerby github commenters. |
|
@smoogipoo Yes, last I've heard and with the implementation trials I have followed, this PR was the goal for some time now About the discussion (usually we would do this in Discussions though) : So as for the 'hidden accuracy' statement, it had been proposed, but understandably it's not straightforward to implement even though it would have made the solution work. Any reasonable new input to expand the realm of possibilities would be welcome, but in this thread I don't see something new (feel free to expand this in Discussions, if the team is looking for potentially more solutions ?) |
|
@Adrriii There's another path to that blue one, which is moving the 305 score to the combo portion. Do you remember anything about that? |
|
@smoogipoo I don't exactly. Sorry it has been a while and I'm not part of the players who really mind combo |
|
I just went through poll results and past discussions, and I don't think that was brought up (explains why it doesn't show on the map). So I don't have any info. Personally, I fail to see any direct drawbacks without seeing different scores with these parameters. My only reserve is that, by doing this, the ratio part of accuracy would be held into the combo portion. Again I don't see what are the direct drawbacks of that if scores still make sense afterwards. |
|
Where are we with this PR because it's unclear? Is anything changing, or is this ready to review? |
|
@smoogipoo maybe that's your call ? (no idea if that was brought up elsewhere, I just hope this PR doesn't die ;) ) |
Resolves #24238
I thought we already had an osu-queue-score-statistics command to recalculate this, but I can't find it. Will need to make one.