Tropical × Service - Editor's Notes and More

Hello. This is miyaryu.

This article is the editor's notes for my work published in June 2024 (← way too long ago!!!!!!!!!!!), or rather, it's a chimera article packed with otaku talk, trivia, and a guide on how I made it. It's extremely long (15,000 characters!!!), so please feel free to jump to the sections you want to read using the table of contents.“Tropical × Service”

Also, if you haven't watched this work yet, please watch it before reading!!! (And if you have watched it, please watch it again if you'd like!)

0. First of all

I deeply apologize for the significant delay in publishing this article. After the release in June, I had a busy time with personal matters, other productions, entrance exams, and moving, and now that things have finally settled down enough to make some nearly a year later, I have finally managed to find some spare time to resume writing. Everything is new and difficult, but I'm managing to do well.

Since my motivation for production has been building up due to various things like Na-Tro and M2, I thought now was my chance to let it all out, so I'm working on several personal projects. Because of my personality of being interested in everything, I can't devote all my resources to just OtoMADs, and my production pace is inevitably slow, but I would appreciate it if you could continue to bear with me.

1. Motivation

First, I'd like to talk about the personal circumstances that led to the creation of my work.

The period from around March to Golden Week in 2024 was perhaps the busiest time of my life, and I really had no time at all to make OtoMADs. After being freed from days of being chased 24 hours a day by a to-do list of over 10 items, and with the collaboration work settling down, mid-May was the first time I was free in a long time. Considering that I had to start studying for entrance exams from the summer, the next few months were the only time left for me, a fleeting fever time. It's ironically similar to my current situation after finishing my exams and moving.

And then, I suddenly remembered an OtoMAD for “May I Help You?” that had been sitting in limbo because the dialogue matching wasn't finished. I had been working on it a long time ago using string-telephone (The Demon Girl Next Door), and I thought I might as well finish it now. However, that memory connects with the excitement of the anime “Tropical-Rouge! Precure” that I had finished watching a little while ago. A bright and comical musical style, and a song structure that makes it easy to pack in basic anime information (character introductions, short stories, etc.). They are a perfect match. The moment I thought of this pairing, an image of the finished product came flooding into my head with incredible force. I just had to make it. For about a week after that, I spent all day thinking about this work.

A slight digression. I started watching the “Precure” series at the beginning of 2024. At the time, I was going through what I personally called a renaissance, a movement to re-collect content I hadn't experienced during my childhood; for example, this was also when I started watching Disney movies properly. I watched Precure out of curiosity as part of that, but before I knew it, I had been dragged into a place I couldn't escape from. It couldn't be helped.
However, the most direct trigger for my interest in Nichiasa was actually mi-ru's PriMagi MAD, “Miruki no Cho-Cho-Ho-Cho”. I believe I first saw it in a Top 10 list, and I loved it so much that I became interested in the original work and surrounding content. I am still happily watching “Himitsu no AiPri: Ring Arc”.

Returning to the main topic. I will now explain the anime "Tropical-Rouge! Precure" lightlyheavily. The protagonist is Manatsu Natsuumi, a first-year middle school student who moved from a remote island to Aozora City. On her first day, she meets Laura Apollodoros Hyginus La Mer (hereinafter
Manatsu Natsuumi. On her first day, she meets Laura Apollodoros Hyginus La Mer (hereinafter Laura), a mermaid who was a candidate for queen of the ocean kingdom, Grand Ocean, and also encounters the evil people who came to steal people's motivation power, along with the monsters they created, the Yaraneeda. Since Manatsu is the "protagonist", she gathers her courage to stand up to the monster to save Laura, who was about to be attacked. Then, a mysterious light appears, she receives a transformation item from Laura, and transforms into Cure Summer! She defeats the monster. From here, the story of the two (and their delightful friends) begins. It has many pop and funny developments centered on gags, like a so-called "slice-of-life" series, yet it is a great work that faces profound life philosophy head-on and presents a solution through Manatsu's words and actions. I was watching it during the aforementioned period when I was very busy, and it became a great emotional support for me. If you're busy, don't watch anime And, as I mentioned earlier, the story progresses with laughter as its axis, which is precisely why the

growth, and the message behind the story surrounding the destiny of Laura, who is a mermaid, are immeasurable. As the story progresses, it gradually becomes clear that the accumulation of casual daily life has unknowingly transformed into irreplaceable friendship. In the midst of this, what befalls the five is a conclusion that is just like the story of "The Little Mermaid": sad, fleeting, and infinitely beautiful. The final episode is seriously, truly amazing, so I hope you will watch it. I do, anyway.

From around here, there is a bit of spoiler warning. The biggest keyword of this work is "do what is most important right now." This is a motto inherited from Manatsu's father (who used to live with Manatsu on a remote island; Manatsu left her father to live with her mother (and later Laura) in Aozora City, where there is a middle school), which means to not be needlessly trapped by the past or future, but to think about what should be done "now", and to take action immediately while that "important thing" is still remembered, while the "important thing" still exists. It is closely involved from the beginning to the end of this work. Between the past and the future~♪

This is different from hedonism (the director said it was different). It is more about thinking about the present with the premise of the future in mind. In other words, if the happiness right in front of you will disappear someday, it's best to enjoy it to the fullest while you can. Following this line of thinking, even the most unpleasant things can be sublimated into a valuable “experience”. This idea also serves as the work's own solution to the serious problem of the lifespan difference between humans and mermaids, as well as to nihilism (I believe works like 'The Demon Girl Next Door' and 'Miss Kobayashi's Dragon Maid' had similar aspects). Now that I have left my parents' home and moved to Tokyo, I feel the importance of these words deeply (not in a regretful sense).

Of course, Manatsu is not a perfect human being. The villain of this work is named “The Witch of Delays”, but even Manatsu puts off unpleasant homework by saying things like, “The most important thing right now is to slack off~.” However, including that, this work depicts it as a lovable “present” (of course, she does study properly in the end). It avoids the excessive moralizing (preachiness) that works for children often fall into, and you can see that it is produced with the interests of the viewers at their past, present, and future stages equally in mind. It doesn't chase after momentary happiness, nor does it encourage self-sacrifice or enduring hardship; I think it is a wonderful work created with the wish for people to truly live happily. I think it is very good that Precure (especially recently) conveys this (perhaps slightly idealistic?) life philosophy on Sunday mornings: first, cherish your own happiness above all else, and if that premise simultaneously leads to everyone's happiness, that's super happy, right? And in fact, I have been saved by it more than a little.

By the way, it is said that the closer you go to a “tropical” climate, i.e., the tropics, the smaller the changes in the natural environment due to the seasons become, and thus the concept of time gradually weakens. This tendency is particularly pronounced in tropical rainforests because the difference in brightness between day and night is small, and it is said that some ethnic groups living there do not recognize the flow of time—past or future—at all (naturally, they have no words to express it). This could be called the ultimate in “living in the present” (even if it deviates slightly from the work's argument), and this might be the reference for the character Manatsu. I don't know, though.

Oh, as a side note, the Tropical-Rouge Precure collaboration episode of Aiseki Shokudo (10/26/2021, (probably) available on U-NEXT) is incredibly funny and recommended. You can enjoy it even if you haven't watched the anime.

2. Concept/Audio

2-1. Concept

Sorry for the long-windedness. It's finally time to talk about OtoMADs. Next is the OtoMAD talk~♪

The biggest concept of this work, in a word, was “evangelism”. To explain to people who don't know Tropical-Rouge at all while making it fun that there is this very interesting work, these characters appear in it, they each have these thoughts, and this atmosphere is the charm! It sounds quite arrogant when written out, but the point is to express the atmosphere of the entire anime and the charm of Tropical-Rouge as much as possible within the OtoMAD format. Incidentally, I started production with the assumption that this would be my last personal work before entrance exams, so I intended to release all the skills I currently have as a culmination of my activities so far. Therefore (or rather, this is just because I am severely lacking in the ability to think of new things), I have incorporated expressions similar to my past works throughout. (← Last year's me wrote this, but I have no idea what part they are referring to. It's a mystery.)

Based on the above concept and the trends of other MADs of the same song, I decided to make it a work that focuses particularly on Vocaloid-style voice synthesis and visual expression. In other words, I started making it without a plan without thinking about matching the dialogue at all, so I would be tormented by it later. Reflection.

2-2. Instrumental

The first thing I started with was the production of the instrumental source. The reason is because I can reuse the 183 PF and it's the easiest. So, I wanted to post the original song and the source with the dialogue/vocals muted... but the PF was corrupted when my PC broke and I bought a new one. Sobbing. I'm really sorry...

Actually, I didn't put much effort into it. Well, it's not a YTPMV, so I had the lazy thought that it would be fine if I just made the sounds I wanted people to hear. Actually, if you turn up the volume of the original song, you can surprisingly fake it. Also, transcribing the end (the Hidekazu Tanaka-esque chord progression zone) was super hard. I definitely made a lot of mistakes.

2-3. Vocaloid-style synthesis

The next thing I started working on was manual Vocaloid. This song really shines when you add powerful manual vocals that ride the sense of openness in the chorus.
This time, I had the five members of the Tropical Club sing in unison. The effort required was, of course, five times as much. It was tough, but since the other audio sources cover up the rough parts that stand out when listening to a solo, it sounds much better.I highly recommend choral singing.
By the way, here are the pre-panning and pre-mixed audio sources for each character. Listening to them like this, I think you can see how powerful choral singing and VSTs are.

I always follow the flow below when creating.

I will explain them in order.

1. Arrange sounds in REAPER
This is the first and by far the most important step. How well you can extract good phonemes without compromising here is the turning point of your fate.

First, if you have character songs (or even if not, singing voices), using them will give you a significant advantage. Singing voices have clearer pronunciation than speaking voices, and the pitch fluctuations are smaller, so the percentage that can be used as material is higher, and the sense of discomfort during processing is greatly reduced. It is even better if the vocal range is the same. Moreover, since they include professional mixing and mastering, the voices cut through easily. Well, it is only natural that reconstructing singing voices to make them sing sounds better than stretching speaking voices to force them to sing.

By the way, if there is harmony in a character song and you cannot extract a single note just by vocal separation, you can use the UVR karaoke mode to isolate only the main vocals. It sometimes fails, but it is very useful.

Once the material is ready, it is time to choose the phonemes. With guts, loop the audio source as many times as needed and find the sound you want. It is (apparently) convenient to transcribe the entire thing in hiragana so you can search it. I have never done that, though.

There is only one criterion for choosing phonemes: "the one with the smallest difference from the ideal". Specifically,

I am considering these conditions. I will explain them in order.

- Small pitch difference
VocalShifter is excellent at natural pitch changes for human voices, but even so, there are limits to its capabilities. If you shift the pitch too much, it inevitably causes adverse effects, such as the voice quality changing or the vowels becoming crushed. Basically, it is good if you can keep it within ±4 to 5 semitones. In addition to that, it is even better if you can pay attention to the vertical relationship with the preceding and following notes (since the resonance changes quite a bit between +4 and -4, it is probably better not to have them too close together), but in the end, I think there is an aspect of luck involved.

・Appropriate singing mood
Choose a singing voice that matches the atmosphere you want to create. For example, if you use a quiet, mysterious voice (like one derived from a B-melody or a drop chorus) in a section you want to build up, it won't harmonize. Naturally, it is also important to maintain consistency with the surrounding sounds. While this is often difficult, it might be better to take source material only from the chorus itself once you have decided on the chorus.

・Vowels are "the same"
"The same" here is meant in a phonological sense. In other words, it is not enough for them to simply be the same "a" in the lyrics. Vowels change in a gradient depending on tongue position and lip shape; even for the same "a" sound, there are various versions, such as those leaning toward "e" or "o" depending on the situation. If you listen to singing very carefully, you should be able to find several places where the vowel pronunciation is subtly different from everyday conversation. You can just fix this if you hear something that feels off, but it is good to be able to recognize when a vowel is the cause of that discomfort. By the way, this can be explained with basic knowledge of phonetics, but I will omit that here as it would take too long.

Instead, please read this article. It is the article that got me interested in linguistics, and it holds a very significant place for me. It is very educational.

・Able to hear consonants and vowels
The important thing here is not "correct" but "able to hear". Minor compromises, such as substituting a "ka" sound for a "ga" sound or forcing consonants and vowels together, will be buried in the flow of other sounds. I believe that if you are going to try to match the lyrics strictly, you should prioritize the aforementioned pitch and mood. In the end, the only thing you can trust is your own (or someone else's) ears. If it sounds good, it's good, and if it feels off, you can just improve it. That's all there is to it. This is also explained in detail and in an easy-to-understand way in the article mentioned above.

I will arrange the phonemes selected with these four points in mind to match the rhythm. I will also list a few things to be careful about when doing so.

・Vowel elision
i/u sounds that are sandwiched between voiceless consonants (s, k, t, p, etc.) or at the end of a word may not be pronounced. For example, when saying "sukidesu" (I like you), I think most people actually pronounce it as "skides" rather than "sukidesu". That's it. If you don't understand, try saying it while holding your hand to your throat. You will notice that your throat does not vibrate during the two "su" sounds. If you don't listen carefully to the original song and correctly analyze where vowels are elided or not, it will sound
strange. For example, if you place a vowel in the "su" part in the example above, I think the audio will sound awkward. Even with original lyrics, try singing them yourself to see where it is natural for vowels to be elided.

・Diphthongs
This will be one of the biggest hurdles in arranging phonemes. For example, if you want to make it sing "nai" (nai), if you divide it into na/i as written, it will sound unnatural. Here, you need to treat the whole "nai" as one syllable (explaining the definition of a syllable would take too long, so I'll omit it; how to divide sounds is enough of an understanding) and connect the a and i manually. If there is already voice material where "ai" is connected, it is best to use that. If not, overlap the two phonemes and force them to connect. At this time, the pitch of the phonemes needs to be (almost) the same, so take the vowels from parts with as close a pitch as possible, and finally adjust them with something like soloist.

There are also many other things to pay attention to, such as the length of sound elongation and timing discrepancies caused by consonants. However, the point is just to listen carefully to the original song's vocals and reproduce them as faithfully as possible. What you should trust is your own ears. It might be true that you need to train those ears. Knowledge of phonetics is quite useful for that, so it might be good to study it.

② Stretching sounds and matching pitch in VocalShifter
Once you have finished arranging the sounds, export them and load them into VocalShifter.
I perform the stretching process here. It stretches the sound more cleanly than REAPER, so although it is a bit tedious, I endure it. Using the Time tab, I stretch only the vowel components while being careful not to crush the consonants that follow immediately.

After that, I match the pitch of each phoneme. I followed a procedure like this.

I will explain them in order.

i. Roughly match the pitch by parallel shifting
The reason I use parallel shifting at this point is to avoid unnecessarily distorting the shape of the original material. For example, if you completely ignore the fact that the pitch swells upward in the middle and change it to a pitch that swells downward, it inevitably creates an unnatural feeling (a processed feel). I am sure you have all experienced this problem when trying to synthesize sounds with large pitch changes.

By the way, nowadays I find it easier and prefer to copy and paste the original song's pitch first and then adjust it to meet the above conditions, but either way, I do not use averaging to match the pitch. That is because it is actually strange for a human to be perfectly on pitch when singing. In fact, if you load human singing into VocalShifter, I don't think there is a single sound where the pitch is kept perfectly steady. It fluctuates significantly up and down, but if you take the overall average, it is roughly the intended pitch. The human brain processes things quite roughly and loosely in that regard, and rather than pitch accuracy, how you fluctuate the pitch within a phoneme becomes important as an expression of subtle nuances.

ii. Connect with adjacent sounds → iii. Adjust areas that feel unnatural
Of course, just using the parallel shift of the original material (or copying and pasting the original song's pitch) will not work well. For example, connecting to the next sound with (in principle) a smooth curve, adding vibrato to long tones, or correcting parts where you feel the pitch is off. While observing the expressions of both the original material's pronunciation and the original song's pitch, I search for a pitch curve that I feel is natural and easy to listen to.

iv. Add accents and nuances

It changes quite a bit. It is important. It is also a place where you can express the character's individuality, so let's be creative. Instead of just pursuing "singing skill," it would be wonderful if you could use your imagination to think, "I bet this character has this kind of habit when they sing."

③ Sound creation/mixing on REAPER
Now, finally, the mix. It is almost a victory lap. While looking forward to the finished product, I just stuff in effects randomly. No, wait, random is not good. This time, I also tried panning each character left and right.

With this, it is finally complete. My management of vowels was sloppy, which is a point of reflection. But I feel like I made the best human-made cover I have ever done. When it was finished, the sense of accomplishment was at its peak 🐢.

2-4 Dialogue matching

Finally, I will add the dialogue (which I had been putting off). I didn't have any material at all, so I had a very hard time.

・Listening to it again, I feel like I featured Kururun too much. But I like her, so it can't be helped.

・Most of the materials used are from the first and last episodes. That is too extreme. Well, these two episodes were the ones that presented the relationship between Manatsu and Laura and the overall outline of the work most clearly, so...

・La Mer was the last additional Cure to appear, so it would be natural to place her self-introduction last, but my desire to place Manatsu and Laura side-by-side won out. I also feel that their narrative positions are largely divided into these two groups of 2 and 3. Also, after watching the final episode, just seeing Manatsu and Laura talking happily makes me want to cry.

・I like the way the sound is abruptly cut off during the "What are you saying? Human." part. I'm happy that I was able to assemble a cinematic soundscape in the MAD that was conscious of space and emotion.

・The reason Minori's lines sound strange is because I used a direct recording. I seriously had nothing else... it was a last resort.

・There was a plan to focus on Laura's first transformation episode during the B-melody, but due to material constraints, it ended up in its current configuration. Either way, the structure of transforming and entering the chorus was already set.

・"That was pretty cool, Manatsu!" ← I absolutely love this ending. I think I was able to follow the delicate changes in the song's expression.

2-5 Complete!

With this, the audio is complete. I think I spent about 12 hours on this part (plus maybe 10 hours on the 183 I reused?). My memory is extremely hazy, though. By the way, the number of layers is about 80, including muted ones. By far the most I've ever used.

3. Video

Most of the composition was decided while thinking about the audio and video as a set, so from here on, it's just the work ofgiving it shape. That's the hardest part, though.

3-1. Overall

First, let me list the things I thought about as a whole.

・Production time was about30 hours? Probably. I did my best for me.

・Regarding camera work, I was strongly conscious of "being aware of howforce is applied" to the video, as mentioned inthis article by Sake-Shoyu. I still keep this in mind when creating, and honestly, I have to admit that this work still lacked enough consideration when I think about it now. My regret is that the transitions were too one-note.

・I used plenty of gradients. This time I used Aviutl's4-color gradient, but honestly, it didn't reach a quality I was satisfied with (for example, when creating a gradient between two colors with distant hues, there was a problem where the saturation was lost because it passed through low-saturation colors by connecting them with a straight line on the color solid). Besides this, curved shapes and complex animations might beorganic expressions that can be calledAviutl's weak point. Including this, I've recently started to feel the limits of Aviutl little by little. I might consider switching to AE after I buy a desktop PC (it's true that blaming the software doesn't start anything). By the way, I've become able to use Illustrator and Premiere Pro at a minimum level.

・I wanted to adjust the amount of editing for each scene with the mood in mind, but it didn't go very well. It's such a bad habit. I want to keep in mind at all times thatunnecessary editing ruins the work.

・I learnedAdobe Illustrator, so I incorporated it a little into this video production. Since then, I've been increasing what I can do little by little, but while it's super tedious, it greatly expands what you can do. AE definitely has higher compatibility, but for Aviutl users who want to raise the level of their work, it might be a good idea to try installing Illustrator first.

・I borrowed some materials from the anime'sofficial website. This method is very useful because it is guaranteed that the atmosphere will match. Recommended.

・I regret a little that the transitions were too one-note. I want to increase my arsenal.

That's all for the general review. From here on, I will explain by part.

3-2. Intro

・The logo was made in Illustrator. It's just a combination of rectangles and fan shapes. There's nothing difficult about it, but it was quite tedious and took about 40 minutes.

・Regarding the top and bottom frames, the clouds and beach in the background were borrowed from the official website. I packed the foreground with silhouettes of tropical motifs. The result is good, but honestly, it's debatable whether it is 100%in harmonywith this scene... It feels like I couldn't output exactly what I had imagined.

・For this part with Manatsu and Sango, I wanted to recreate the feeling of a handheld camera. This is the limit.

・The context is a bit weak, and I feel like the video ended up being just something that looks the part. That is something I regret.

3-3. A-melody

・This is the self-introduction zone. I divided it into two-measure segments. For the first half, I cut out key moments from each character's transformation scene and their finishing poses, connecting them by adding transitions that match each movement. For the second half, I created a unified format.

・For the text and background colors, I blended three image colors for each Cure. Gradients are difficult, aren't they?

・The font for the first half is Logo Jr Black. Personally, I don't like to use it actively because I feel that using it indiscriminately creates an excessive sense of casualness and dominates the atmosphere of the video, but here, it harmonized the best.

・The character name assets are from the official website. I applied a glossy finish to the 'Annyantroman' font. I can't deny that there is a sense of color mismatch with this cut, and it might have been better to create them myself. ・The font is Rodin. I think it matches the resort-like tropical atmosphere well. I remember using this font a lot in my school trip (to Okinawa) guidebook.

・conscious of the space, and for transitions, I moved each object individually (e.g., keeping the movement of background objects subtle). Camera control is useful when this is tedious, but it comes with limitations (e.g., you can't use clipping on objects above), so it's a difficult problem. The bad parts of Aviutl~~~

・The concentration line-like effect behind this cut involved a slightly tedious process. First, I prepared what is introduced in this article by Imo Tart (https://immortalt.hatenablog.com/entry/2020/11/07/211822). I placed that in the scene, applied mosaic, and then binarized it to make it look dot-like. I loaded this as two scene objects (for two colors). I also used it in my work 'Dodgeball's Sand Canyon,' and I feel it has become one of the techniques that symbolizes my style. I can't put it into words, but it's convenient and good to use.Maybe if you can't put it into words, you shouldn't use it.

3-4. B-melody

・Adjusting the amount of editing was difficult. Or rather, I failed.

・I wanted to lower the tension of the editing here a bit to prevent the viewer from feeling suffocated, but in the end, while prioritizing the balance and rhythm with other cuts, the screen ended up being quite busy. A bad habit!

・I can't do text motion. I'm already at my limit with even something this simple. I think I still don't understand at all why cool motions are cool. A year has passed and I've gotten a little better, but I still can't see the backs of the amazing people. After all, I don't feel like I can reproduce the subliminal-like direction built frame-by-frame to influence people's unconsciousness using the logic I have acquired so far. I wonder if I should analyze good videos frame by frame.

・I'm also bad at color grading. I feel that color requires the most individual 'sense,' that is, an eye to judge for yourself what is good and what is bad, but I don't have that sense.

・This time, I incorporated tone curves into color correction for the first time.

It is effective when you want to strongly bring out a certain atmosphere, but it was difficult because if you don't handle it carefully, it seems like it would lead directly to what is called 'Aviutl-ness.' But honestly, I feel like this is also just Aviutl-like.In the first place, trying to do color grading in Aviutl is a mistake, isn't it?

3-5. Chorus

・I took screenshots of this whale one by one and cut it out. The trick to OtoMADs is not to mind the work of cutting out dozens of images, no matter how little benefit there is. If you can use AI, use it. If you can't, honestly move your hands. By the way, I don't use GIMP, but this smartphone app (https://apps.apple.com/jp/app/%E8%83%8C%E6%99%AF%E9%80%8F%E6%98%8E%E5%8C%96/id926035241). The advantage is that you can do it with intuitive operations while lying down or on the move, and the disadvantage is that the image is compressed to a certain size (about 1600 square).

・I worried quite a bit about the first transition of the chorus. I had to make sure not to dampen the sense of openness of the audio with the video. In the end, I don't feel like it went very well.

・I made the background myself, referring to the original special move background. It has a structure of gradient -> polka dots -> splashes -> concentration lines -> particles. The processing is deathly heavy!

・The circle behind the text was conscious of bubbles. I'm doing something strange using text objects to make them appear one by one. Despite putting a lot of effort into it, the effect is weak. Specifically, I followed the steps below.

It doesn't really harmonize with the background in terms of color, and I regret that I could have put more effort into that. The idea was good, though.

・It's frustrating because when memory is low, scene objects often fail to load correctly during output. Sometimes, merging multiple scenes into one surprisingly solves this. When things are really tough, you have no choice but to use multi-pass output (outputting scenes to video files and replacing them) or split output (outputting by part and joining them later) (I have wasted an infinite amount of time stubbornly trying to avoid this).

・This is the ending sequence. I think it's amazing that the Precure series has been using such high-quality 3D animation for its EDs for many years. You should watch the ED for the new series 'You and Idol Precure♪'; the color design in the chorus is on an otherworldly level.

・What's spinning here is a background with a diagonal clipping applied. I recently learned that if you set the width value to a negative number, it crops the center. Convenient.

・The cutout of the bubbles here was achieved through brute force by placing circles to match the movement of each bubble and cutting them out with clipping. I really wanted to create a sense of depth on the screen. It's definitely not a cost-effective task, but a creator is someone who does whatever it takes to achieve the expression they want. That's half the reason; the other half is that I was in a brain-dead state and didn't want to think of an alternative.

・I created this text using glow, inspired by those glittery pens you see with purikura or those things where you move fireworks around to make letters. It's cute, stylish, and I'm pretty fond of it.

3-6. Outro

・The three cuts that follow from here went extremely well. 'Borsalino × Service' is my own development of that same part.

・If there's something I'm lacking, it's composite skills. Everyone is so good at using things like tritone or binarization to create ideal color spaces, and I admire that. I, on the other hand, sometimes even slack off on color correction...

・The surrounding objects were all placed in the space by sheer manual effort. It took a huge amount of time. Honestly, I'm just using passion to make it look decent here, so I have nothing else to say...

4. Conclusion

This concludes this article. Thank you so much for reading such amateur writing. I would be happy if this has been of any benefit to you at all.

Finally, I would like to express my utmost gratitude to those who mentioned this on social media or in comments, and to those who selected it in voting events like the Top 10. I intend to make OtoMADs for my own self-satisfaction, but even so, seeing that my work has reached someone's heart makes me feel glad to be alive.

Well then! Let's get tropical today as well!!!