when will the new ranking system be put up? or is there no specific date?

When it's done.

Tom said one week ago he was planning on implementing the current algorithm to makes taiko pp run. So it should be delivered soon

Just lost around 8000 of pp. Why me?

Haibanemu wrote:

Just lost around 8000 of pp. Why me?

You're not the only one.

pp is being recalculated, hang in there.

All hail Strangeprogram !

Umm... that was not intended. *goes and fixed weighting of lower accuracy... drastically*

Ok so the one who FCed this will just become insta #1 because he can spam his keyboard ? (Or maybe using 2 keyboard) . That's typically what we wanted to prevent by ignoring converted maps, because stuff like that is going to be overrated as hell. (But yeah, lower the gain of bad accuracy on insane map is already a good start)

The second big issue is that it will bring even more people cheating/using more or less legit tricks to do that kind of stuff just to earn pps.

what happen with the rank? i lost a lot of pp and gain a 3% of acc, is this working as intended?

i dont think it was intended so we should see a fix soon enough

I have a question, with this current implementation of pp, are converted maps weighed differently compared to taiko-specific still?

yes^

anyway, i feel like DTHR should be weighed much more especially on higher OD
my OD12 score on seasons of asia is beyond human like (considering accuracy gods such as zx didn't even get anything close to my acc) and yet it's nowhere mentioned in my BP
97.5% on OD12 is nearly impossible for a legitimate player

i have converted maps on my profile as "Top Ranks/Best Performance" when it was used to be only specifics maps and some of the best players have this change too, what thats mean?

To develop a little about my best performances, I was kinda expecting to see chipscape/image material and stuff like that in my BP so that's okay. But there is some weird result out there which I'd like to point out :

Bring Me The Horizon - Anthem : I don't really get it. Ok this map is long, but it's far from being a crazy map. I didn't even use HR on this and my acc is below 99%

DJ YOSHITAKA - Evans : This is the one that less belongs here. It's a common oni, it's not even hard to acc (OD4 HP4, Settings lower than what we usually put) and I have an average accuracy on it regarding the setting (99.2). I have a lots of HR SS on maps that are, imo, harder or at least equivalent difficulty-wise like ende, corrosion, etc...

Yousei Teikoku - Hades: The Rise : Quite the same as Anthem, even if it's an harder map.

Overall, I think that, apart from the excepted maps, my real best performance aren't there. I have made quite a lot of SS HR or enven DT on 170+BPM maps which are not here and far harder than 80% of my current best perf. (SigSig https://osu.ppy.sh/b/99114&m=1, E.T (190BPM DT) https://osu.ppy.sh/b/111287&m=1, etc..)

Okay, let's start with one super obvious example of a map that breaks the algorithm: Kanzen Kankaku Dreamer
Both the Oni and Muzukashii are *ridiculously* overrated. I can't even begin to describe how stupidly high the pp yield on this is.

Let me illustrate:
kevin3333g (best accuracy player ever): #1 performance KKD Oni; #2 performance KKD Muzukashii
asgardv (top tier player): #1 performance KKD Oni; #2 performance KKD Muzukashii
TKSalt (top tier player): #1 performance KKD Oni (this is a 4x100 nomod score btw, wtf?!)
The list goes on and on and on. Just finishing this pretty much guarantees it will be your #1 top performance.
Just for reference, I got my HR SS on that map literally first try, so it's not a hard map by any means - just long, but not even *that* long by accuracy players' standards.

Now for the other maps in my top performance list:

AugoEidEs - This was a good score back when I got it 8 months ago, but nowadays it's really not that impressive anymore. The patterns are simple and repetitive, my accuracy is subpar compared to pretty much everything I consider my true top scores. The BPM honestly isn't even that high by taiko standards (207 BPM), so the only real difficulty here are length and SV.

rog-unlimitation - This is actually a good score, nice job algorithm. Doesn't really deserve to be all the way up at #3, but I certainly won't complain if this stays in my top 10.

Rengetsu Ouka - Uh, why exactly is this here? I got it literally first try, not that hard. I mean, it's not a bad score or anything, but the BPM is not high enough to make it actually difficult. My accuracy isn't too hot either.

Mythologia's End - Wat. This score isn't impressive at all. Mediocre accuracy, I even got misses. Even 8 months ago when I got this score, I wan't pleased with it. Only difficulty is map length.

Son of Sun (Extended Mix) - Well... I personally don't really like this score, but I can see why your algorithm would consider it a good one. I can accept this one (reluctantly).

Power of Dream (Night Fever Refix) - Shitty accuracy, but at least the map is kinda tricky. Same as for Son of Sun; I don't really agree, but I can accept it.

Magia (TV Size) - Accuracy is literally the only difficulty here, and I have like 50 scores that I'd consider better accuracy performances.

Seikan Hikou - Another score I got on my very first try. I ragequit after getting it because my accuracy (only real difficulty on this map) was so terrible, so it kinda hurts my feelings seeing it in my top performance list haha

Rainbow Dash Likes Girls (Stay Gay Pony Girl) - Honestly not all that hard, and I missed on it. Really outdated score as well. Oh well, at least it's harder than most maps on this list, so I'll take it.


The feeling I get is that high BPM is overrated in a weird way. Playing anything below 200 BPM doesn't seem to be worth much (unless it has 1/6 streams and OD9.8 like Power of Dream). Yes, high BPM is generally harder, but 220~ isn't as difficult in taiko as it is in standard. I have no issue with weighing 270 or 300 BPM scores really highly, but right now the scales seem to be tipped way too far in favor of 200-220 BPM maps. A better way to put it might be "lower BPM is underrated".
That being said, I don't fully understand the speed weight either - my FC on Forgotten for example is a lot more difficult in terms of speed than anything from my current top performance list. While I agree that it's not an outstanding score due to mediocre accuracy and repetitive/simple patterns, it's still harder than 90% of what the algorithm considers my top plays (the 10% being rog-unlimitation).
My accuracy plays don't seem to get any love. Stuff like Endless Tears..., Manima ni, Munchy Monk Circus, Anison Sprint (Anime Song Medley) and others are (in my opinion) a lot more impressive performances that all of those mediocre 200-ish BPM scores.

Having A with DT should always be better than SS HR/HD

EBAWER123 wrote:

Having A with DT should always be better than SS HR/HD

Quick, someone rank a 100 BPM OD8 maps so my DT A can beat all those pesky no-skill HR SS /s

Luna wrote:

EBAWER123 wrote:

Having A with DT should always be better than SS HR/HD

Quick, someone rank a 100 BPM OD8 maps so my DT A can beat all those pesky no-skill HR SS /s

Even though you think you are sarcastic that is actually 100% correct =)

TIL 150 BPM A is harder than OD10 SS

https://osu.ppy.sh/b/61676&m=1 BPM: 114 OD: 5 - No dt records =)
https://osu.ppy.sh/b/63993&m=1 BPM: 143 OD: 5 - 1 dt recor

Surely HR must be so much harder

You're cherry picking maps and generalizing that DT must always be harder.
Give me an hour and I'll get like 20-30 A scores on 150 BPM (will have to intentionally miss to get As actually). I doubt you can even produce a single OD10 SS in that time. Feel free to prove me wrong. But then again, you'd need to get a ton of OD10 SS in that time to actually prove me wrong.

Note that the hit window for 300s of OD6+DT is the same as OD10. DT does make the OD skyrocket in Taiko, contrary to standard.
Edit: My mistake, it's actually OD6.66 + DT which is equivalent to OD10. Sorry.

OD7 DT (OD5 HRDT) is equivalent to OD10
OD6 DT is easier

Luna wrote:

OD7 DT (OD5 HRDT) is equivalent to OD10
OD6 DT is easier

My mistake, it's actually OD6.66 + DT which is equivalent to OD10. Sorry. OD7 + DT would be slightly harder.

Ebawer this is just stupid, I can have A on a tons of Taiko diff even knowing I'm way far from FC'ing, the system will say that this A rank is better than my HR SS even if i'm 50k points behind it. And most of time those A ranks took me just one try or two. You can't really say it's harder or easier, it just require a different skill. Atm players that can full alternate barely DT and clear everything with a bad accuracy their ranks will be higher for the pp system than those who have good accuracy on a lower bpm.

Converted maps needs to be weighed much less, especially spam maps.

Sorry to the mentioned guy, but https://osu.ppy.sh/u/1451311#_general

Look at his best performances.


all the maps under "spam" are maps with impossible 1/8 streams - which are barely playable without HT or EZ.

same for this guy: https://osu.ppy.sh/u/1506011, his BP almost only consist of spam maps, here having only ONE taiko diff. other people have insane scores on taiko diffs that are really overshadowed by spammy converts which got magical fcs like strangeprogram. this is what we wanted to avoid, actually.

Just for the record, I'm not saying HR SS is always better than DT A - a DT A on Big Black is harder than a HR SS would be for example. I'm just saying that generalization ("DT A is always better than HR/HD SS") is ridiculous.

And I agree on the spam conversion part...

_Gezo_ wrote:

Converted maps needs to be weighed much less, especially spam maps.

Sorry to the mentioned guy, but https://osu.ppy.sh/u/1451311#_general

Look at his best performances.

SPOILER

1. spam
2. SPAAAAAAMMMMMMMM
3. spam
4. spam
5. spam
6. idk didn't play
7, 8. taiko so it's ok
9. idk
10. spam

all the maps under "spam" are maps with impossible 1/8 streams - which are barely playable without HT or EZ.

same for this guy: https://osu.ppy.sh/u/1506011, his BP almost only consist of spam maps, here having only ONE taiko diff. other people have insane scores on taiko diffs that are really overshadowed by spammy converts which got magical fcs like strangeprogram. this is what we wanted to avoid, actually.

His spam scores are pretty damn ridiculous. I've never seen a human come even close to that speed so far, so I'm not exactly sure how to view those.

The general opinion I get here is that I should penalize misses more, weight combo higher and consider accuracy more, is that correct?
(Those changes would penalize this guy's spam scores, too, as a nice side effect.)

If you use two keyboards at the same time, you can reach speeds at that level and even beyond on single color streams. It's not easy to coordinate your fingers so as not to hit two of the same key at the same time, but it can absoultely be done. I don't know if the players in question used this trick, but it's still a possibility, so please just decrease the weight of conversions by a significant margin ;w;

they did, both midnait and whosthebox can't play for shit xD same as kawhy, one kb is more than enough, lol, you can either use 3 keys for each colour or use 2 hands for spamming 2 buttons, alternating left-righ hand, kawhy had a guide how to do it but I cba to look for it.

@Tom no need to alter the rankings, let those people be.

So does this system take a detour from peppy's goal in standard for accuracy weighting? From my new top ranks, it seems the pp is based less on the score itself and more on the difficulty of the map itself, with less weighting on how I actually played lol

Can vouch for the BPM 200~230 being weighted too heavily.

Looking at my own best performances, apart from that A on augoeides (unlucky miss), other As shouldn't be there. Those plays on those maps were damn terrible and I feel bad for being that terrible, and they are weighted much higher than some other maps on the list.

But I can understand this in a way, as a still relatively new taiko player, I don't have the ability to play high bpm, so that's that I guess. Anything above 190 is actually pretty damn hard for me, hell even some lower bpm maps are really hard for me. But I have to say my scores on BRA, Hades and Ernst are pretty damn terrible, with multiple misses.
I can probably guess why Hades is being valued this much (was trying to talk about it before), but I can't get why BRA is. Apart from the speed component and one stream which starts with a kat on a blue tick, the map is just repetition of the same patterns over and over (though the SV could be tricky maybe, don't think it's THAT hard though) and I think it's being valued too much. My score over there sucks and it still beats some more impressive scores I did (imo).
Also about Ernst, the map introduces slightly longer patterns than your normal 3~5 note patterns, but it isn't really that hard to be honest. My performance there was terrible and is being weighted more than some other more impressive scores like Shawarma HR.
Oh yeah, Destrudo Oni being that high, uhhh yeah that's really wrong. That map is a lower tier 8* Oni and most of the taiko people can FC it first try with great acc. Should be faaaaaaar lower, same with Uchida Maaya - Tsubomi Yume Miru Rhapsodia ~Alma no Michibiki~ [Inner Oni]. my own maps derp

The list looks far better than before though. Was full of converts or some really easy Onis haha.

whosthebox play one hand kb one hand mouse IIRC. still that converted maps are pretty hard to do i dont know how that could be easier than a inner oni

After looking through a lot of top players' performance lists, I've come to the conclusion that the rhythm complexity factor might be overrated a bit. It was intended to boost the difficulty or maps with mixed 1/4 1/3 1/6 streams and similar patterns, but right now I get the impression that maps with just one or two stray 1/3 patterns are considered significantly harder than regular 1/2 1/4 maps. Maybe it's just coincidence and the real issue lies elsewhere, but that's what i gathered so far.

Luna wrote:

After looking through a lot of top players' performance lists, I've come to the conclusion that the rhythm complexity factor might be overrated a bit. It was intended to boost the difficulty or maps with mixed 1/4 1/3 1/6 streams and similar patterns, but right now I get the impression that maps with just one or two stray 1/3 patterns are considered significantly harder than regular 1/2 1/4 maps. Maybe it's just coincidence and the real issue lies elsewhere, but that's what i gathered so far.

It'd be better to look through the difficulty list I posted in this thread for checking this.

http://osu.ppy.sh/b/90616?m=1 this map being on my top ranks despite a miss gives me the feeling that the high SV section at the very end is skewing the maps rating a bit, especially when a much more difficult to FC song like http://osu.ppy.sh/b/284037 isn't on my top 10 yet Taiko Time is.

Luna wrote:

After looking through a lot of top players' performance lists, I've come to the conclusion that the rhythm complexity factor might be overrated a bit. It was intended to boost the difficulty or maps with mixed 1/4 1/3 1/6 streams and similar patterns, but right now I get the impression that maps with just one or two stray 1/3 patterns are considered significantly harder than regular 1/2 1/4 maps. Maybe it's just coincidence and the real issue lies elsewhere, but that's what i gathered so far.

Factors for a better rank:
Taiko Mode: Death Streams should not be considered to better rank
Standard Mode: Separate notes very fast possibly
CTB: Many sprints for better rank ... no have idea
mania: ... only acc ._.

Also, My rank from #27 to #312 ... why? Q_Q

Tasha wrote:

http://osu.ppy.sh/b/90616?m=1 this map being on my top ranks despite a miss gives me the feeling that the high SV section at the very end is skewing the maps rating a bit, especially when a much more difficult to FC song like http://osu.ppy.sh/b/284037 isn't on my top 10 yet Taiko Time is.

Combo isn't weighted as heavily in Taiko as it is in standard at the moment. Not exactly sure how much I should change that.