This is a BSS beatmap submission. Click here to view full beatmap information.
User avatar
Beatmap Nominator
456 posts
Offline
SV (Slider Velocity) Information for someone who doesn't have conviction about average


First, I used ms (millisecond) to average SV instead of snap (1/4, 1/8, etc) actually, it's my second experimental attempt about how to use SV.
Using snap to SV is generalized ms ratio so ms will work same with it tho.

let first SV duration = a (ms), first SV = x, second SV duration = b (ms), second SV = y. It should be

a + b = Real length (ms)
ax + by = Calculated length (ms)

If Real length is same with Calculated length, we can tell "It's averaged SV".

1. x = 5.00, y = 0.50 (Real length : 137ms)
a + b = 137ms
5.00a + 0.50b = 137ms

Therefore, a = 15.2222ms, b = 121.7778ms. (I round off the numbers to five decimal places.)
Put a, b to Calculated length equation
(5.00 × 15.2222ms) + (0.50 × 121.7778ms) = 76.111ms + 60.8889ms = 136.9999ms.
136.9999ms ÷ 137ms = 0.99999927...., It's 1 actually so we can tell it's averaged.
We can apply this equation to 'x = 0.50, y = 5.00', 'Real length : (137 ÷ 2)ms', 'Real length : (137 × 2)ms'.


2. x = 3.00, y = 0.33 (Real length : 137ms)
a + b = 137ms
3.00a + 0.33b = 137ms

Therefore, a = 34.3783ms, b = 102.6217ms.
Put a, b to Calculated length equation
(3.00 × 34.3783ms) + (0.33 × 102.6217ms) = 103.1349ms + 33.8651ms = 137ms.
137ms ÷ 137ms = 1, so we can tell it's averaged.
We can apply this equation to 'x = 0.33, y = 3.00', 'Real length : (137 ÷ 2)ms'.


3. 0.40 ~ 1.60 speedups
I used agka's sv tools, each SV has same duration, so you can know it's averaged easily because (0.40 + 1.60) ÷ 2 = 1.


My whole SV equations are totally three, so you can check it's averaged.
Hope you understand my not enough explanation :V

PS : edited part 2 sv, updated
User avatar
Rhythm Incarnate
638 posts
Offline
No kudosu yet.
수학이여~
User avatar
Beatmap Nominator
456 posts
Offline
Wonki wrote:
수학이여~

노가다여~
User avatar
Beat Clicker
71 posts
Offline
No kudosu yet.
Hello Garalulu!

Although overall structure seems great but since SVs are bit questionable, I decided to put some random suggestions here.
I agree that everyone can get different feelings from same sound and can express with different SV styles, but your SVs don't match with music since it seem to be too strong.

1. x = 5.00, y = 0.50 (Real length : 137ms)
Why are SVs at 00:00:983 - // 00:01:392 - // 00:02:483 - // 00:10:120 - are all the same?
SV at 00:00:983 - is for vocal, 00:01:392 - is for background piano sound which lasts longer than 00:02:483 - and new part (only instrumental) is starting at 00:10:120 - and you used all the same SVs (If I guessed right)

Also, in my opinion 5.0x is not a value we use for ordinary sounds (which is usually considered as teleportation SVs), and there does not exist strong sounds at the starting section of this map.

So what I recommend is,

Using smooth/weak effects for sounds like 00:01:392 -

00:01:392 - 2.7x / 00:01:409 - 1.6x / 00:01:427 - 1.0x / 00:01:444 - 0.7x / 00:01:461 - 0.5x / 00:01:529 - 1.0x

and use shorter (n bit stronger) effects for sounds like 00:02:483 -

00:02:483 - 3.0x / 00:02:500 - 1.4x / 00:02:517 - 0.6x / 00:02:620 - 1.0x

00:09:574 - to 00:10:103 - seems to be fine since drum sounds are short and strong, (fine to keep 5.0x / 0.5x SV for these sounds)
but I don't agree with 00:00:983 - those vocal sounds should be expressed with same intensity of SVs that are used for drum sounds.

Maybe 00:00:983 - 3.8x / 00:01:017 - 0.6x / 00:01:120 - 1.0x would feel better. (or you could just put same effect that is used 00:02:483 - here)

Or you could make average scroll speed to 0.9x for starting section since sounds are weaker than other parts.
Average SVs are regulated to be 0.9x
00:01:392 - 2.5x / 00:01:409 - 1.4x / 00:01:427 - 0.9x / 00:01:444 - 0.6x / 00:01:461 - 0.45x / 00:01:529 - 0.9x
00:02:483 - 2.5x / 00:02:500 - 1.1x / 00:02:517 - 0.6x / 00:02:620 - 0.9x
00:00:983 - 3.35x / 00:01:017 - 0.55x / 00:01:120 - 0.9x

Also I'm not sure about putting another SVs 00:05:415 - here or seperate LN cuz lyrics are ko ko r- ko ko r-o wa~
I won't make similar suggestions for 00:10:120 - and other parts since you have enough skills to handle it, but I hope you also make other parts smoother. (not for drum sounds they are fine)



2. x = 3.00, y = 0.33 (Real length : 137ms)
I don't agree with using SVs at timing where the sounds have no significant difference like 00:40:597 -
So for this SV, I recommend you to use,
00:40:392 - 2.5x / 00:40:427 - 1.3x / 00:40:461 - 0.7x / 00:40:495 - 0.5x / 00:40:648 - 2.5x / 00:40:665 - 1.0x
These values regulate both parts'(00:40:392 - to 00:40:597 - // 00:40:597 - to 00:40:665 - ) average scroll speed to 1.0x


3. 0.40 ~ 1.60 speedups
Some players say that their PC is not good enough and if there are too much (or too dense) SV points in short section, their PC lags and made them to fail.
So in my opinion it would be better to pick 4 to 8 timing points (that time difference of each SVs are the same) and put values like
0.4x 0.8x 1.2x 1.6x or
0.4x 0.6x 0.8x .... 1.6x

Not a problem for my PC tho


Again, patterns are fine but SV seems to be need moding at least 1 or 2 times by modders who are good at handling SVs.

No kudosu and hope this helps! :)

PS: 앞에 븐첵 이야기가 있어서 영어로 적었습니다
User avatar
Spinner Sage
130 posts
Offline
No kudosu yet.
Hello OwO

I just want to hear your opinion.
In the most sections, the SV does not change its value to 5.0 0.5, which seems to be a concept.
I think this is not good. I think Otakujanai is think the same thing too.

Think about this after reading Otakujanai's post.

(저는 영어 모딩을 컨셉으로 잡았습니다 좋지 않으신거 같으면 답글달아주세요~)
User avatar
Beatmap Nominator
456 posts
Offline
Otakujanai wrote:
Hello Garalulu!

Although overall structure seems great but since SVs are bit questionable, I decided to put some random suggestions here.
I agree that everyone can get different feelings from same sound and can express with different SV styles, but your SVs don't match with music since it seem to be too strong. I think it fits what this music wanna express like chaos so not too strong I tho.

1. x = 5.00, y = 0.50 (Real length : 137ms)
Why are SVs at 00:00:983 - // 00:01:392 - // 00:02:483 - // 00:10:120 - are all the same?
SV at 00:00:983 - is for vocal, 00:01:392 - is for background piano sound which lasts longer than 00:02:483 - and new part (only instrumental) is starting at 00:10:120 - and you used all the same SVs (If I guessed right)

Also, in my opinion 5.0x is not a value we use for ordinary sounds (which is usually considered as teleportation SVs), and there does not exist strong sounds at the starting section of this map.

So what I recommend is,

Using smooth/weak effects for sounds like 00:01:392 -

00:01:392 - 2.7x / 00:01:409 - 1.6x / 00:01:427 - 1.0x / 00:01:444 - 0.7x / 00:01:461 - 0.5x / 00:01:529 - 1.0x

and use shorter (n bit stronger) effects for sounds like 00:02:483 -

00:02:483 - 3.0x / 00:02:500 - 1.4x / 00:02:517 - 0.6x / 00:02:620 - 1.0x

00:09:574 - to 00:10:103 - seems to be fine since drum sounds are short and strong, (fine to keep 5.0x / 0.5x SV for these sounds)
but I don't agree with 00:00:983 - those vocal sounds should be expressed with same intensity of SVs that are used for drum sounds.

Maybe 00:00:983 - 3.8x / 00:01:017 - 0.6x / 00:01:120 - 1.0x would feel better. (or you could just put same effect that is used 00:02:483 - here)

Or you could make average scroll speed to 0.9x for starting section since sounds are weaker than other parts.
Average SVs are regulated to be 0.9x
00:01:392 - 2.5x / 00:01:409 - 1.4x / 00:01:427 - 0.9x / 00:01:444 - 0.6x / 00:01:461 - 0.45x / 00:01:529 - 0.9x
00:02:483 - 2.5x / 00:02:500 - 1.1x / 00:02:517 - 0.6x / 00:02:620 - 0.9x
00:00:983 - 3.35x / 00:01:017 - 0.55x / 00:01:120 - 0.9x

Also I'm not sure about putting another SVs 00:05:415 - here or seperate LN cuz lyrics are ko ko r- ko ko r-o wa~
I won't make similar suggestions for 00:10:120 - and other parts since you have enough skills to handle it, but I hope you also make other parts smoother. (not for drum sounds they are fine)

What I want to show is repetitive sound structure like the broken old tapes, not the instrument or vocal's sence of dynamics. The music is composed of piece of sound(not just vocal or drum, the music 'renai circulation') and expressed like defective jukebox. so I use same sv at all of what I want to add, You maybe have feeling of sv in mind, but it's just my concept yeah I don't wanna use various sv for this.



2. x = 3.00, y = 0.33 (Real length : 137ms)
I don't agree with using SVs at timing where the sounds have no significant difference like 00:40:597 -
So for this SV, I recommend you to use,
00:40:392 - 2.5x / 00:40:427 - 1.3x / 00:40:461 - 0.7x / 00:40:495 - 0.5x / 00:40:648 - 2.5x / 00:40:665 - 1.0x
These values regulate both parts'(00:40:392 - to 00:40:597 - // 00:40:597 - to 00:40:665 - ) average scroll speed to 1.0x

Exactly it's fine, thx for suggestion but I want to keep this.


3. 0.40 ~ 1.60 speedups
Some players say that their PC is not good enough and if there are too much (or too dense) SV points in short section, their PC lags and made them to fail.
So in my opinion it would be better to pick 4 to 8 timing points (that time difference of each SVs are the same) and put values like
0.4x 0.8x 1.2x 1.6x or
0.4x 0.6x 0.8x .... 1.6x

Not a problem for my PC tho

I play osu without graphic card. Just old AMD CPU but I don't think that I have a problem because of amount of SVs, maybe someone have a problem, not just sv, but osu itself.


Again, patterns are fine but SV seems to be need moding at least 1 or 2 times by modders who are good at handling SVs.

No kudosu and hope this helps! :)

PS: 앞에 븐첵 이야기가 있어서 영어로 적었습니다


Lastopia wrote:
Hello OwO

I just want to hear your opinion.
In the most sections, the SV does not change its value to 5.0 0.5, which seems to be a concept.
I think this is not good. I think Otakujanai is think the same thing too.

Think about this after reading Otakujanai's post.

(저는 영어 모딩을 컨셉으로 잡았습니다 좋지 않으신거 같으면 답글달아주세요~)

I said about this issue at otakujanai's mod, thx for your opinion.

표현하고자 하시는 변속을 직접 적용해보고 생각해보았으나 역시 저는 기존 입장을 고수하고 싶습니다. 왜냐면 위에서도 말했듯이 단지 보컬의 목소리, 드럼의 세기에 따라 강약을 조절하고 싶은 것이 아니라 조각난 노래들이 어떻게 반복되며 재생되는지를 표현하고 싶었습니다. 그래서 그런 부분에서는 모두 같은 크기의 변속을 넣은 거구요. 곡의 컨셉이라고 이해해 주셨으면 좋겠습니다. 변속이 강하다고 생각하실 수 있지만 저는 이 정도가 음악이 표현하고자 하는 방향성과 어울린다고 봅니다. 변속의 강함과 롱놋을 의식해서 OD는 7.5로 조정한거구요. 의견 감사합니다!

Thx for your suggestions!
User avatar
Rhythm Incarnate
1,206 posts
Offline
No kudosu yet.
just checked the map im fine with the svs, most of the chunk of the difficulty comes from processing patterns between the x5 sv jumps which is fairly easy to pick up.
User avatar
Beatmap Nominator
632 posts
Offline
Earned 1 kudosu.
hi hi heres heres my my mod mod

1|2|3|4

softhitnormal has a really long empty tail, should definitely fix it

use this bg instead http://animewallpaperstock.com/wallpape ... ko0060.jpg

juan would like this map methinks

koi koi

00:15:574 (15574|1,15574|3,15574|2,15574|0) - I dont think these quads are really necessary, they feel really strong when the sound isnt really that different. removing notes like 00:15:574 (15574|1) - also removes some of the shields which makes it play smoother

00:28:938 (28938|2,28938|1,29483|1,29483|3,29483|0) - general note count inconsistencies. Same sound, different note amount
while we're on it, these reds 00:29:347 (29347|1,29347|0) - shouldnt be doubles, they're even weaker than the blue ticks imo

00:49:392 (49392|3) - my god
00:58:802 (58802|2) - MY GOD

I get the intention of this section but a lot of notes feel mapped to nothing like 00:57:983 (57983|0). This should be a 1/2 LN but it keeps going to match the theme which idk if im ok with. This can be extended to other sections as well like the kiai and first part where the are LNs that are just too long like 01:22:392 (82392|3) - where the last part is clearly not mapped to anything. since the song goes silent except for the vocal in col 2

hard

01:03:711 (63711|2) - having sounds like these as LNs is very questionable, gonna need a justification for these

01:07:938 (67938|0,68211|0) - same sound (literally kappa), different LN length

LN lengths are definitely the biggest problem in this set, they're not even consistent across difficulties

Image

norm

possible spread issues Image
User avatar
Beatmap Nominator
456 posts
Offline
Litharrale wrote:
hi hi heres heres my my mod mod

1|2|3|4

softhitnormal has a really long empty tail, should definitely fix it Fixed

use this bg instead http://animewallpaperstock.com/wallpape ... ko0060.jpg backfire will give me something so bg will be replaced :p

juan would like this map methinks Hope so :D

koi koi

00:15:574 (15574|1,15574|3,15574|2,15574|0) - I dont think these quads are really necessary, they feel really strong when the sound isnt really that different. Oki
removing notes like 00:15:574 (15574|1) - also removes some of the shields which makes it play smoother I agree. Deleted one

00:28:938 (28938|2,28938|1,29483|1,29483|3,29483|0) - general note count inconsistencies. Same sound, different note amount I decrease one count when the LN is placed.
When the LN isn't placed 00:29:756 (29756|0,29756|3,29892|3,29892|1,30029|0,30029|1,30029|3) -
When the LN is placed 00:28:665 (28665|1,28665|0,28802|3,28802|0,28938|1,28938|2) -

while we're on it, these reds 00:29:347 (29347|1,29347|0) - shouldnt be doubles, they're even weaker than the blue ticks imo 00:29:347 (29347|0,29347|1,29415|2) - they are same sound tho, but I put double at red line only because I don't wanna make double - double - triple(it makes awkward jack), so only emphasis the red line.

00:49:392 (49392|3) - my god main drum for double, sub drum for single
00:58:802 (58802|2) - MY GOD it makes same effect

I get the intention of this section but a lot of notes feel mapped to nothing like 00:57:983 (57983|0). This should be a 1/2 LN but it keeps going to match the theme which idk if im ok with. This can be extended to other sections as well like the kiai and first part where the are LNs that are just too long like 01:22:392 (82392|3) - where the last part is clearly not mapped to anything. since the song goes silent except for the vocal in col 2 First, 00:57:983 (57983|0) and 01:22:392 (82392|3) - are pretty different concept.
1. 00:57:983 (57983|0)
I put double at main drum like 00:48:302 (48302|0,48302|3) - , single at sub drum like 00:48:438 (48438|2) - .
but they are intentionally stretched, because I wanna emphasis the vocal+drum's endless repeated chaos. First part 00:45:029 - 00:53:756 - and second part 00:53:756 - 01:02:483 - looks similar, second part LN is same with first part but more-stretched because background melody is added. I wanna make a blend of repeated medley to put stretched LN.

2. 01:22:392 (82392|3) -
it's same concept with 00:01:392 - 00:10:120 - , I put double for melody but for drum I divide one like 01:20:211 (80211|1,80483|1) -


hard

01:03:711 (63711|2) - having sounds like these as LNs is very questionable, gonna need a justification for these you're right. I changed this LN to just a single note

01:07:938 (67938|0,68211|0) - same sound (literally kappa), different LN length changed the length, thx!

LN lengths are definitely the biggest problem in this set, they're not even consistent across difficulties

Image
But it's not same diff, just for space management. exactly you can't make three LN's ending 01:20:211 (80211|0,80483|1,80620|2) - and two LN's starting 01:20:756 (80756|1,80756|0) - to put same time (Koi Koi diff that you show)
norm

possible spread issues https://i.imgur.com/853dPof.png wow actually, I add one note at 01:18:438 -


Thx for modding!
User avatar
Beatmap Nominator
632 posts
Offline
No kudosu yet.
Garalulu wrote:
But it's not same diff, just for space management. exactly you can't make three LN's ending


Well that's the point. A lot of these LNs aren't really mapped to anything. The start is but then they just kind of end wherever rather than being mapped to a specific sound in the music. You can't have notes that aren't completely mapped to a sound, that borders on ghost notes/dump mapping.
User avatar
Beatmap Nominator
456 posts
Offline
Litharrale wrote:
Garalulu wrote:
But it's not same diff, just for space management. exactly you can't make three LN's ending


Well that's the point. A lot of these LNs aren't really mapped to anything. The start is but then they just kind of end wherever rather than being mapped to a specific sound in the music. You can't have notes that aren't completely mapped to a sound, that borders on ghost notes/dump mapping.

Well, actually I didn't put ghost note anything, LNs have their meaning of existence. Different LNs length on same position at two diff (hard, koi) not mean it's inconsistency, it's flexibility about each situation. I do keep consistency at same diff actually. each diff follow each rule, I can assert that they did. If you wanna put all of sound to map and same shape without any situation, it will be awkward structure you know. LNs are placed with melody, drum, melody sound at my map. Even though it looks excessive, it's for possibility of LN usage and connection of each part, re-analysis of music at my opinion. plz understand about it, thank you!
User avatar
Beatmap Nominator
632 posts
Offline
No kudosu yet.
I'm not talking about between diffs (although that is an issue)

I'm talking about most of the LNs in the set that don't make sense.

An illustrated example:

This highlighted LN here 00:55:392 (55392|3) -
Image

(video for people reading this who dont want to open the map https://puu.sh/yhTPB/c1e22186bc.mp4)

It is 2.75 beats long and starts on a clap and vocal sound accompanied by two SNs to show the clap.

I'm assuming the LN is following the vocal because there isnt really much else for it to follow. Standard logic would say to end the LN when the vocal ends which is correct. The vocal ends at the start of the very next LN here 00:55:665 (55665|2) - but the LN keeps continuing. Is this a bad thing? not at all, having an LN continue as another syllable of the vocal is mapped is a technique that's used and that makes sense.

Problem is the second syllable ends here 00:55:802 (55802|1) . You could argue that it ends at the white tick if you really stretched it but both LNs then continue beyond that as well. At this point we're going into uncharted territory where these LNs only exist as patterns rather than something mapped to the music.

They don't even end at the same spot so using the argument of "mapping separate syllables of a word then ending at the same spot" doesn't work

Image

So if they're not mapped to the music, why do they end at all? why don't they keep going forever? Because new LNs start and we need to make room for them.

This is not ok. More than 50% of this LN isn't mapped to anything and that might as well fall under the category of "ghost notes".

The LN used as an example 00:55:392 (55392|3) - should end after one beat, 2 beats if you want to stretch it. Any further is ghost noting.

The map is full of them as well. I get what you're going for but I wouldn't consider it rankable.
User avatar
Beatmap Nominator
456 posts
Offline
Litharrale wrote:
I'm not talking about between diffs (although that is an issue)

I'm talking about most of the LNs in the set that don't make sense.

An illustrated example:

This highlighted LN here 00:55:392 (55392|3) -
Image

(video for people reading this who dont want to open the map https://puu.sh/yhTPB/c1e22186bc.mp4)

It is 2.75 beats long and starts on a clap and vocal sound accompanied by two SNs to show the clap.

I'm assuming the LN is following the vocal because there isnt really much else for it to follow. Standard logic would say to end the LN when the vocal ends which is correct. The vocal ends at the start of the very next LN here 00:55:665 (55665|2) - but the LN keeps continuing. Is this a bad thing? not at all, having an LN continue as another syllable of the vocal is mapped is a technique that's used and that makes sense.

Problem is the second syllable ends here 00:55:802 (55802|1) . You could argue that it ends at the white tick if you really stretched it but both LNs then continue beyond that as well. At this point we're going into uncharted territory where these LNs only exist as patterns rather than something mapped to the music.

They don't even end at the same spot so using the argument of "mapping separate syllables of a word then ending at the same spot" doesn't work

Image

So if they're not mapped to the music, why do they end at all? why don't they keep going forever? Because new LNs start and we need to make room for them.

This is not ok. More than 50% of this LN isn't mapped to anything and that might as well fall under the category of "ghost notes".

The LN used as an example 00:55:392 (55392|3) - should end after one beat, 2 beats if you want to stretch it. Any further is ghost noting.

The map is full of them as well. I get what you're going for but I wouldn't consider it rankable.

Maybe you don't take LN maps enough to understand this, will tell about LNs if you really want
I hope not
User avatar
Beatmap Nominator
632 posts
Offline
No kudosu yet.
Garalulu wrote:
Maybe you don't take LN maps enough to understand this, will tell about LNs if you really want
I hope not


Code:
-       _   -
User avatar
Global Moderator
1,307 posts
Offline
Earned 1 kudosu.
I see LN discussions so I will give my opinion on them (my poor ears ;_;)

As for the consistency between diffs I don't think it is a problem at all. That's kinda the point of making something harder and its being different, structured in harder ways.

If I have to be concerned by something it would be how the SVs make up for the patterns which in general feel aggresive but playing them are actually not as complicated as they look. (for example 00:02:688 (2688|2,2688|0) - this one being a double instead of 00:02:756 (2756|1) - which in theory it breaks the consistency of the pattern a bit but again, it could be justified with proper SV usage)

As in general speaking for LN releases I want to sort them in two different types, intentionally hard or intentionally easier. The case of this map features mostly intentionally easier releases due to the way it is constructed. With so many SVs having wonki (pun not intended) releases would make it really complicated to follow so having the release timings attached to the next LN press for instances would make up for a more playable pattern (and in my eyes, it is completely rankable, you can check the kiais of FIRST as well).

Other cases of LN sort of simplification is doing stuff like this 00:11:892 (11892|2,11961|1) - forcing the LN releases to be both consistent at the same time but also representing the pitch / sound change. This way given the pitch or sound change is not as relevant as it could be for other instances using techniques like this one makes sense for me as well. Same applies here 00:45:029 (45029|0,45302|1) - but also with SV justification.

00:45:029 (45029|0,45302|1) - For this section what it matters is the LNs that represents similar instruments (or voice in this case) to keep up with a logical pattern (which is trill oriented shields) but also keeping up consistent lenghts within each other (all of them are 3/4s long).

00:57:983 - Here it is basically full inverse, its intentional and its fine as well I would say. My only concern would be where the inverse actually starts which imo should be 00:58:120 - here with the main long sax going long (which also leads me to believe that 00:57:983 (57983|0,59074|3,60165|1) - this ones are unsnapped by 1/2)

01:04:665 (64665|2,64665|3,64802|2,64938|3,65074|2,65211|3,65279|2) - This pattern shape is not that bad but man making it right hand is a bit of a pain. I would split it between both hands.

01:19:938 - You are evil and I hope you know it. I love this. This plays susprisingly well with the SVs and I hope you dont change it. Even when it breaks a bit the consistency in terms of density the pattern transition makes up for the change in my opinion. Specially considering the SVs.


Anyways, I just wanted to point those things out to like hopefuly clarify a bit my opinion on this one and hopefully help the mapper in order to push this forward (this song is really annoying aaadfafasf)

Feel free to call me if you are in doubt (Lith or Lulu whoever needs it) for anything here. I would gladly offer my help or at least provide my opinion in any particular part you feel its not right.
This is a BSS beatmap submission. Click here to view full beatmap information.
Reply 62 posts jump | prev  1, 2, 3, 4, 5  next

Users browsing this forum: None


Jump to: