Vrmd wrote:
Hello Endaris!
I'd like to ask you a couple of questions about the things you pointed out when modding my map ( viewtopic.php?p=4861496#p4861496 )
Endaris wrote:
Set a lower AR, this is uncomfortable to read. I would recommend any value between 8 and 8,3. Right now you can see only 1/1 ahead which makes it rather difficult to identify different patterns. I tried 8,5 and it still feels too fast(I can provide math for why it shouldn't be AR9 if you insist)
I am not sure I understand what you mean with being able to see only 1/1 ahead. My personal impression is that AR9 currently gives the best balance between seeing enough notes ahead to understand what's going to happen and not having too many approaching circles on the screen at the same time as to be confusing, specially in some sections like 02:43:862 (1) or 03:07:281 (1); when I lower the AR to 8, I get more easily confused by the jump 03:08:055 (5,6,7) for example (my brain has trouble deciding if the jump goes right left or left right) while with AR9 it feels more clear I have to actually go right-left.
I am curious as to a mathematical basis for choosing the AR, so if you can elaborate on that it'd be grateful
Endaris wrote:
00:12:506 (7,1,2,3,4,5,6,1,2,3,4,5,6,7,1,2,3,4,5,6,7,8,1,2,3,4,5,6,1,2,3,1,2,3,4,5,6,1,2,3,4,5,6,7,8,9,10,1,2,3,4,5,6,7,8,9,10,1,2,3,4,5,6,7,8,1,2,3,4,5,6,7,1,2,3,4,5,6,7,8,1) - The structure of this entire part is pretty meh.
I think this whole section unnecessarily difficult to play because it is not obvious which parts of the music you're valueing over the others..... (rest of quote omitted for brevity)
I have a dilemma about your explanation about this section, and the whole map in general. I agree on basically everything you mention here: about drums going basically on in the white and red ticks, about strings being the predominant sound in this section, and that it feels natural that following the strings given that the drumming will be covered by the beats that will happen on white and red ticks.
As a matter of fact, the example mapping you placed as a picture in your post looks a lot like how my mapping of that section looked a few versions back. My problem is, if I go that way when mapping, it increases heavily the quantity of long sliders (you added a couple ones in your example), which in turn makes the map feel more boring to play given its difficulty is Insane.
Since I had a lot of those sliders after my initial mapping attempt, I ended up trying to balance between following the drums more and using these long sliders that emphasize strings or vocal sections more sparsely - which is why it has that inconsistent effect that you mention. It seems I can't decide between following the drums and the strings because that's exactly what's happening - I am having a hard time deciding exactly that. If I follow the drums too much I feel a predominant sound is being left out, and if I just follow the strings (or the vocals in other parts of the song) it starts feeling boring and kind of slow paced.
What would you recommend I could do to properly balance these concerns?
Regarding your other modding comments, I agree with them and will work on fixing those.
Thank you again for your all your help!
Sincerely,
Vrmd
Hi, thanks for your detailed reply!
I wish more people shared their concerns about a mod in such an elaborate way.
Okay, so AR will be first.
The things you have to consider when picking an AR are BPM and which intervals between notes are used in your map(and partially how) and which ones are not.
Your map has a speed of 155bpm.
In terms of intervals it uses a lot of 1/2 spacings, some 1/4 and 1/1 spacings and a few 3/2 spacings.
For AR9 the approachtime is 600ms. However a player cannot immediately recognize the circle. As circles fade in(=they're effectively not visible in the very first moment of appearance) it is valid assumption to say that a player needs a tiny part of that 600ms to even recognize it(10-30ms, depends on the player too).
Now 1/1 spacings appear relatively frequently on your map and 1/1 means 387ms difference on the timeline.
On a spot like 01:10:378 (1,2) - this means the following:
You don't see the next circle until you almost finished the slider. For me the circle becomes visible enough for easy recognition at around 01:10:862 - and even quickwitted people won't notice it before 01:10:789 -
This means that the player is basically "afloat" for the majority of that slider because he doesn't know where to go next, there is barely a maximum of 1 object on-screen during that time and when you start moving for the appearing circle you won't know where to move after that - it will become apparent somewhere on the way but it is uncomfortable as stated.
From my point of view it is desirable if the player can read 1 object ahead at any point of time because otherwise the game feels more like reacting to the circles appearing instead of seeing a rhythm developing in front of you and following it through.
Now this was an example for 1/1 spacing but you do have some 3/2 spacings too and these are almost a full 600ms(581ms) apart from each other which causes the ongoing rhythm to reset in a way because the playtime kind of "ends"(as there are no visible objects you could move for) and restarts a splitsecond after. That's a terrible thing to happen.
So that is why you absolutely should use an AR lower than AR9. 8,7 already extends the approachtime to 645ms which already kind of fixes the spots but imo there are good reasons why an even lower AR is better.
(A different argument is that AR doesn't influence the intensity of the song at all)
To go on, almost all of your 1/4 patterns are stacked/have a constant and low spacing, meaning that they don't contribute to the actual density and reading difficulty of the song. In the moment the player identifies them as stacked 1/4 he will treat them as ONE gameobject. . 02:43:862 (1,2,3,4,5) - is about the only real exception and that's also why you should rather change it to have it in a line with the rest of the song.
And by the way identifying 1/4 patterns becomes harder at high AR as a short approach time will make circles appear rather one by one instead of a stack right away(mainreason why many people including me can't play AR10)
I already talked about 3/2. Your map is mainly based on 1/2 and each 1/2 is 193ms apart from each other
That means with AR8.7 you can see the next 3 1/2 spaced circles in the moment you click a circle. Effectively there will be less objects on screen due to sliders reducing that amount.
Now the difference in the amount of objects on screen between 645ms approachtime and 704ms(=AR8.3) approachtime is effectively 0 as the fourth circle will appear 774ms after you clicked the last one.
tl;dr: The average amount of objects on screen is the same for AR8(=750ms), AR8.3(704ms) and AR8.7(645ms) as the important values of increased density are
581ms and
774ms BUT lower AR-values tend to benefits the spots where you use bigger spacings like 3/2 as the relatively difference between seeing the next object 50ms before having to move for it and seeing it 100ms before moving for it is very big and significantly affects the feeling during gameplay. AR8 already feels a bit edgy as the next circle will already appear if you hit a circle late.
That's why AR8.3 would be my preferred choice for the map.
For the other thing:
One reason why the long sliders feel bad is the low SV. I believe that they will be less boring if you increase the SV and the other thing is that you don't have to make them 3/2 like I did in my example, 1/1 is sufficient to give the sound time to reasonate imo.
You can also go for 1.7 or 1.8, would accelerate the sliders even more than my suggested 1.6. As the unpleasant gameplay during the long sliders seems to be mostly caused by a bad SV-choice I would recommend you to try increasing it first and message me again if you're still not happy with the result.
Aside from that you could also force changes of pace through the sections between the long notes, like increasing the spacing on this pattern 00:27:216 (1,2,3,4,5) - and giving it more comfortable angles at the sametime as you would have some designed speedup towards the important sound and then slow down as a relaxing feature once you get there. Giving them a gameplayrelated value by actively manipulating the player's cursorspeed before them(when appropriate) seems to be the most logic choice.
No kd of course, I just wanted to post it here for timestamp-conversion and cause I spent enough time on the post that it's actually worth seeing.