01:26:325 (86325|3,86325|2) - no sound here like at 01:22:725 (82725|1,82725|0) - you could just remove or i think itd look nicer if you made 01:26:250 (86250|1,86250|0) - into a [14] chord and make them ln for the increasing suspense here
01:30:075 (90075|1) etc are intentional for the purpose to represent the vocals rather then the violin, 01:30:975 is removed to have emphasis on the notes 01:31:050 (91050|0,91050|1,91125|3,91125|2) - etc
01:32:250 (92250|2) move this by 1/4 to blue line there sound is on blue tick
after 1:30~ the rest of this part till the technical part is dumped mapped for vocal. 01:32:250 (92250|2) has a better transition into the following part, is louder so this representation is fine.
01:41:775 (101775|3,101793|2,101925|3,101943|2,102975|2,102993|3,103125|2,103143|3,104175|2,104193|3,104325|2,104343|3) - same pattern 3 times in a row? mirror one of these so it doesn't feel repetitive.
01:43:550 (103550|2,103650|2,103725|2,103837|2) - very uncomfortable to hit. sacrifice 01:43:762 (103762|0) - and move 01:43:725 (103725|2) - to col 1
02:07:200 - 02:09:600 this section felt way too empty than what the song's section is trying to cover
perhaps buffing this into a more rice-oriented structure may enmend it (representing the lack of the "violin" melody)
maybe working it out like this
https://osu.ppy.sh/ss/19517574/9110
https://osu.ppy.sh/ss/19517576/bc68
02:36:000 (156000|0,156225|0) - vs 02:36:225 (156225|3,156600|3,156825|3) - the stacking distribution felt a bit uncomfortable (through gameplay) i'd suggest to mirror 02:36:975 (156975|2,157050|3)
04:36:071 (276071|3,276071|0,276115|1,276159|2,276204|0,276248|2,276292|1,276336|0) i guess this is representing the same sound as 04:36:071 (276071|3,276071|0,276115|1,276159|2,276204|0,276248|2,276292|1,276336|0) - or 04:32:542 (272542|3,272601|2,272659|1,272718|0,272777|3,272836|1,272895|2,272954|3) - but in the first timestamp it clearly doesn't show up
just rice-up this section to enmend the hand balance
04:48:418 (288418|3,288525|2,288534|3,288643|2,288653|3,288777|2,288777|3,288865|2,288954|3,289012|2,289071|3) - and 04:59:895 (299895|2,299917|3,300071|2,300150|3,300189|2,300336|2,300358|3,300424|2,300525|3,300645|2,300667|3,300777|3,300777|2,300895|2,300931|3,301020|3,301042|2) - felt too right hand biased, i'd recommend to give the left hand (column 1 and 2) some protagonism too)
(i just found out this is a mirrored structure already, still i think the density intercalation between both pair of columns can improve imo)
removed two notes at 04:48:418 (288418|3,288525|2,288534|3,288643|2,288653|3,288777|2,288777|3,288865|2,288954|3,289012|2,289071|3).
for 04:59:895 (299895|2,299917|3,300071|2,300150|3,300189|2,300336|2,300358|3,300424|2,300525|3,300645|2,300667|3,300777|3,300777|2,300895|2,300931|3,301020|3,301042|2) I think the left and right hand balances out this difficulty because of the mini jacks on the left hand 05:00:228 (300228|1,300314|1,301091|0,301174|0). Both hands are equal because this patterned is mirrored
06:44:188 (404188|2,404206|3,404263|2,404288|0,404288|3,404338|1,404388|2,404438|3,404488|1,404506|0,404563|1,404588|3,404588|0,404638|2,404688|1,404738|0,404788|3,404825|2,404863|1,404900|0,404938|2,404975|3,405013|0,405050|1,405088|3,405125|2,405163|1,405200|0,405238|2,405275|3,405313|0,405350|1) - i feel like the progression can be slower as this felt very sudden by just appearing after a less dense hybrid patterning
06:44:188 (404188|2,404206|3,404263|2,404288|3,404288|0,404338|1,404388|2,404438|3,404488|1,404506|0,404563|1,404588|0,404588|3,404638|2,404688|1,404738|0)
i think you can nerf this part to a progressive rice-to-ln structure avoiding the least of minijacks as shown here 06:44:188 (404188|2,404263|2) - and here 06:44:488 (404488|1,404563|1) - as they are showcased very well already here 06:45:959 (405959|0,406030|0,406102|3,406173|3,406245|1,406316|1,406388|2,406459|2) -
rearrangement suggestion https://osu.ppy.sh/ss/19517616/3fcd
07:04:530 (424530|2,424530|1,424530|3,424673|2,424697|1,424721|0) - reaally subjective but this part could potentially turn into a transition from long note/hybrid-ish to fully rice in a way that the notes represent descendent long notes in terms of length
08:06:122 - 08:07:949 felt veeery empty compared to the following sections; i get that only the percussion is playing but it really decreased it's density a lot - i'd suggest to play with some dump patterning in here or add more graces to intensify a bit more this section
like for example here 08:07:166 https://osu.ppy.sh/ss/19517628/f382
08:15:959 (495959|0,496000|2,496042|0,496084|3,496125|1,496167|3) - i think you can do a similar burst like done in an already ranked map (if this is controversial then it would be better to just take it as a reference) https://osu.ppy.sh/ss/19517631/2b3e