Super Hidamatsuri

It’s been half a year since Majormilk released Super Hidamatsuri, a Hidamari Sketch collaboration 4 years in the making.

Back then, I promised him that we would write about this collab more extensively, as we both had a lot to say. Today is that day.

Chapters

• Chapters
• Timeline of events
  • A love letter to Hidamari Sketch
  • The visual direction
• Assembling the Yamabuki Squad
• Personal experiences
  • MajorMilk
    • Mistakes into Miracles
    • Life is Gay
    • Audiountermensch
    • WW2 Pt. 2
    • The US Postal Office Sucks
    • It’s (Finally) Over
  • CHFR
    • Drafted in the Wide army
    • Working with MJRMLK
    • Working on my part
      • The audio
      • The visuals
      • A sidequest before the finale
    • Final thoughts
• A message from everyone

Timeline of events

A love letter to Hidamari Sketch

The central point of the collab (and what later on caused Majormilk so much grief when it came to finding participants) was the source material: a Kirara series from 2004. We sought to bring in people who all similarly loved Hidamari Sketch, and had, at the very least, seen the entire anime adaptation.

This focus on Hidamari Sketch is not just limited to the source selection, the medley itself is entirely made of openings and character songs from the series.

The visual direction

Classic YTPMV visuals were not encouraged, from the very start, a list of videos was used as this collaboration’s visual guide:

• 【MAD】Paint It Air -ゆゆ式-
• [ MAD ] [ ひだまりスケッチ ] [ ゆのっちが可愛すぎて生きていくのが辛い。]
• YURUYURI’♪♪　(Yuru Yuri × WОRKING’!!　: Op MAD)
• おーぷん☆きゃんばす - HidamaRemix♪

Basically: something colorful and sparkly, with an emphasis on scene composition and absolutely no screenflipping!

Assembling the Yamabuki Squad

The decision of drawing participant portraits was inspired by the credits of the third Non Non Biyori collaboration (which both Majormilk and I took part in), and the stellar job done by the illustrators of the recent Otomad Dream Match.

We were able to get it done thanks to Majormilk himself and his many artist connections.

Personal experiences

MajorMilk

Mistakes into Miracles

While this message might be lost in the sea of collab servers I left ages ago, I vaguely remember telling mords (my co-host at the time) and some other folks that I wanted to organize a collab that wasn’t either Gachimuchi or Inmu, since those had become too stressful for me. A collab that would be a little more self-serving. At the time it was just a vague promise as a result of the direction either collab was going, but it eventually materialized after I started Hidamari Sketch on my birthday.

In the late stages of both gay collabs, my body was stretched to its limit. Christmas season was spent agonizing over late submissions so I stayed up for several nights waiting for last minute submissions. The last week of 2019 had me receiving an average of 4 hours of sleep, mostly involuntary. So what do I do during this tenuous period? I watch a very relaxing and laid back anime known as

Yeah, maybe that wasn’t the greatest idea.

This only made things worse (as much as I loved the story.) Really screwed up my sleep even further. So I had recently heard about Hidamari Sketch during my many Kirara binge watches, so I decided to give it a try. And apologies in advance for writing this out of order, but it’s just the way my mind has placed these events.

During my birthday, I decided to pick up Hidamari. I thought it was kinda cute, but also dull. Was a great way to fix my sleeping schedule back then. The girls were okay, but nothing special. Finished the first season and left it in the Do Back Burner for a while. After my Madoka binge, I had lost all hope in myself but I just knew that Madoka was GOD. Fantastic anime, we don’t talk about Rebellion.

After that I decided I needed a palette cleanser so I decided to pick up HDMR again. For some reason this time the show had blasted me with such warmth and love for the characters that I lost several nights worth of sleep over it, only slowing down when the focus was on Nazuna. Maybe it was an uptick in production quality or something but the anime had drilled itself into my soul. I had no idea this was still possible.

For context, prior to that, the last anime that had a similar effect was Azumanga.

Hard to believe, I know.

But holy cow, that was an experience. I feel like it saved me from completely losing it after the disastrous launches from New Year’s. I was a different person after that, I could not stop gushing over the series. I blasted right through the manga and bombarded my own server’s anime channel with panels and screencaps of Yuno saying the cutest things to Miya.

Not sure how one thing led to another, but a mix of dissatisfaction with the collab scene and my own love for the anime somehow conceived the idea for Super Hidamatsuri.

Life is Gay

I gathered a small band of folks from my server that had also fallen in love with the anime, including the very co-author of this article, and put them in a small channel. I had a very low stakes view of the collab at the time. I originally thought it was going to be one of the 114514 “do whatever” collabs that are all the rage these days. So I had no issue with organizing it on a public server. I contacted my nigga in blue and white, FinnOtaku, and asked him to make sense of all my music picks. I listened to dozens of CD releases and organized them using the following process

And dumped them all on Finn. He was a real sport about it and gave me a few variants of the medley. It quickly delved into a coherent medley. When I organize a medley, I like to have a heavy hand in the composition even if I know jack about music. I try to give enough feedback to let the composer know when things feel “wrong.” I wish this process was more present in other collabs, more often than not you have these hosts give little to no feedback and you end up with the most dull and disjointed composition of all time.

(A bit related, but you can really tell early on when a collab’s a good idea when you’re not dueling over the song picks. I’m talking to you, IWF 2017.)

I showed the medley draft to the folks over at Homo Labo and they liked it. At this point, I realized that this was going to be a more serious collab so I moved all organization over to another server. The population was scant, with maybe a dozen people initially joining the efforts. CH and Darius were the first cannon fodder to be drafted into the collab and somehow lasted the entire process. Slowly but surely, it started filling up a little more with some Korean medley makers joining the fray. From the very start, I knew we were always going to be short on manpower, so I told the composers to start working right away. This was not a democratic process like every other collab I had done in the past. This was a MajorMilk autocracy. You WILL make a banger rendition of Koi-iro Shooting or I will shoot myself.

This was all fine and dandy for most of the first half of the year. Medley was received warmly and discussions were productive. Then I ran into my first of many issues. Oz Han could not muster a satisfactory enough composition for Mebae Drive. For clarification, this was in relation to himself. I heard the previews and I liked them. But you know, artists.

So I got in touch with kotatsu. This fella was a pretty talented music artist that had been following me for a while due to my many Yuyushiki artworks. I got in touch with him and asked him if he could drop in and he accepted. While I don’t have many of the details at hand right now, he left the collab but came back later.

Somehow, the medley was completed.

Audiountermensch

This is where the problems really start. The audio phase was probably the most troubling aspect of the collab. I am mainly a visual person, so I have no auditory skills to speak of. I had to beg and plead with many audio participants to stick around. I served them 5 star course meals (full audio libraries from the show) in an attempt to get them to stay. However, you can’t force a participant to maintain interest in an anime he had little thoughts on. We even had a professional musician on board that offered assistance on some of our parts!

But YTPMVs are quite tricky.

After his first submission, we both realized that it was not quite up to par with what others had in mind. In an effort to facilitate a visualization of a “good” MAD audio, I showed him this

【ゆゆ式】BLU RAY Disc

It sort of helped, he added percs afterwards and submitted as is.

The original band I had spoken about earlier had fallen off the wagon so to speak and left for various reasons. Some involved being disinterested with the final selection of songs or had just ghosted me. Some of these guys were audio so it was extra difficult for me to fill in their works.

At the start of the collab, I had too much optimism for my own good. I let the deadlines stretch up to a month. This was a very free collaboration with no clear deadline and lax checkups, so there was little push on the participants to submit anything. I thought that by gathering the equivalent of Hidamari jihadists, that they would automate everything themselves. But our community has some clear flaws, which are strengths outside of organized efforts. They are very carefree and fickle. We make videos on anything and everything on a whim. As a result of my lax direction, many pushed my collab VERY far down their list of priorities. Some pushed it out entirely.

I lost many early members this way and did not realize it until the second year that this rule had to go. Sometime after New Year’s 2021, I enacted the rule that check ups would be shortened, down from a month to two weeks. This helped a little and I got a few more submissions. I decided to also tune down my rule of only getting participants that had done HDMR videos before. The previous rule was never going to work out, most tagged videos on Nico were over a decade old. The authors had long since retired from production. The ones I did manage to scoop up were on their way out. This resulted in, what I would call “cohorts”. These cohorts would be called in waves to help out the collab, only for all of them to leave shortly after. Entire batches of participants were dropped for very similar reasons at the same time. Mostly due to inactivity.

WW2 Pt. 2

While things were problematic in Super Hidamatsuri, the situation for everyone had become even more dire outside our collaboration. During 2021-2023, slews of international collaborations had gone South and stained the view of foreigners in the Japanese circles. It was harder than ever to gather participants. I could not apologize enough for the actions of my own countrymen other than make some melancholic DMs on Discord. These events caused me to raise my command and forcefully remove a participant. He had been responsible for not only submitting a very half-baked audio (with little regards to pitch shifting, lyrics, or any musical sense.) he had also been responsible for one of the most infamous causes of international miscounduct. Some very brave souls removed this person from power and tried to salvage what damage he had caused on another project.

This was not the end, however.

The fallout of the situation could be felt throughout the year, with constant vague and direct tweets from the Japanese at us. It was a rather embarrassing situation for our community. I was getting sick of dealing with it. Even well after those events, I was working hard to make sure our collab would not fall into the death trap of being overseas project #1145141919810 put in hiatus. Unfortunately, I cannot say the same for all our Japanese participants. Of all the participants I brought in, most of them would mostly waste participant slots as they’d often leave within a short span. They would be way over their heads and unable to conjure up ideas for their parts. I’d get a simple “sumimasen” and a blank slate. Occasionally I’d get nothing.

While I am still truly sorry for the way some of my compatriots acted, I simply could not brush off the typical Japanese attitude I had been experiencing. I was doing everything in my power to make sure the collab was chugging along. Constant sessions with other people on feedback, assistance with material, etc. I was there, acting as a host SHOULD act. But this was not enough. At one point, I came to a realization.

“This is why the Yakuza ask for your pinky.”

Apologies are meaningless in Japan. Especially with a lack of tangible action from the person.
I had the dullest most copypasted DMs of my life during this period. I replied little to nothing to them. I just clicked Remove and sent them on their way, they were of no use. I hated that I had taken this utilitarian mindset, but after seeing it so often, you start to mesh all their messages into a simple neat little box of disappointment.

The US Postal Office Sucks

After this, I decided to lean on foreign participants I was friends with. Surely these guys would be a little more motivated than before? Took my chances with a close friend of mine, he lived in a certain land where the internet is terrible but was quite motivated to watch all of Hidamari Sketch. I was eager to help and personally shipped many materials and even a neat hand drawn Miya on the box. Paid the postage of 120 USD. It arrived a little bit after a month, half complete because the government of that country is full nanny state trogs that consider Hidamari to be actual child pornography or something. The package was half ripped and one of the drives was missing, which had the materials necessary. It was absolutely terrible.

So we tried the next best thing: Pump his phone with enough data to download the full 30GB set. This was a success. In hindsight, I should’ve done this from the beginning.

Unfortunately, this story did not have a happy ending either. The person had many issues with his life and could not muster to finish watching the anime and ultimately ended being terminated from the collab when a new cohort swept through the server once more. During this time, I figured that I was never going to have a complete set of audio makers in time to finish the collab, so I decided to move ahead and make some visuals in advance on parts that I declared to be less sensitive to audio cues. By this point, I had already cleared Sate Sate due to its limited role in the medley. I started to tackle Ume’s section and asked DariusSheeno to help me make a cute Ume model for the part. Uncertain if this experiment would even work, I pressed ahead and built the model little by little.

It worked!

This part was completed and the visuals meshed very well. Probably my most detailed work up to date. The amount of callbacks to Ume’s CDs were tremendous.

It’s (Finally) Over

Despite this, I had made little progress on the audio front. Participants were still missing and nothing was being done. WOOZYWINKLE FROM THE HIDAMARI SKETCH SERVER, once claimed that the collab was dead. It got to me. Enough to start doing some serious scouting around. The collab was reaching its second anniversary and it wasn’t even close to halfway completion. Things were so dire that I was begging CH to fill in the rest of the audio if I were to not find anyone. I knew I was racing against time. The more I let the collaboration drag on, the more probable other factors would knock participants off the project. I considered just paying a professional composer to fill in the blanks of the project, but never pulled the trigger. I did not have enough funds at the time and it was still not certain they would ever actually do it. I already had a musician drop the other parts, why wouldn’t it happen again?

Fast forward, it is now New Year’s 2023. Two years had passed and the collab was in a rocky position. Many “never ever” collabs were starting to come out. The two FAWL collaborations, the KYM collab, even ManWith10Toes had its full audio leaked. This did not bear on my psyche well. So after I got back from my Mexican backwater vacation, I put my foot down and set some very clear deadlines, check ups would be shortened even further. I asked creators with little experience with Hidamari but were motivated friends of mine to help. When smileyshirt started posting his progress on the server, my stress levels started to come down. It was relieving to see familiar faces patching up my crappy project. While there were times when I had to replace members, it was not the overhauls I had in previous years. I was more than happy to oblige.

I finally had a deadline set for audio, end of June.

While most audio participants did meet this date, a handful of them didn’t. Smiley in particular had made a very detailed guide on the samples throughout the anime and it took some time. He worked super hard on it and the end result was beautiful. Definitely worth the wait.

End of the year was fast approaching. By October, we reached a major milestone: the audio had been completed. That thorn on my shoulder had finally been slayed. The result was wonderful.
First night I listened to it, I had such a WIDE grin on my face. It felt like the auditory version of a house you built by yourself. Very homely, comfortable, and warm. All that sweat and tears, all those DMs to CH where I was borderline suicidal, all culminated to this

However, once again outside forces were having a tug on the collab. End of the year collabs are infamous for clogging up everyone’s time. It was having a strain on myself as well. Then, WoozyWinkle brings to my attention that Hidamari Sketch’s 20th anniversary is going to be very soon. In February!

He gives me the idea to delay the collab and save face by moving it to a special date. I immediately take it to vote to the participants and everyone accepts. A huge sigh of relief washes over the server and we all relax for a bit.

Soon after, all art is complete and our trailer goes live on February 1st. I could not sleep afterwards. The culmination of 3 years of work led to this. Everyone KNEW the collab existed outside of desperate cries for help on Twitter. The reception was great and we chugged along.

Finally, two days before the collab was meant to go live. Aya submitted his part.

On February 26, 2024 15:05 UTC, the final part was complete. The collaboration was done. It had not been in vain. All those people that I promised I would send this collab out the door one way or another, would be fulfilled. I had no words.

Only a GIF.

There is little to say after this. I finished the medley PV with everyone’s cute art pieces on their corresponding sections. Winkle was kind enough to lend me his Hidamari merch for the setting and added a nice atmosphere to the whole thing.

I cannot thank enough to all the people who were part of the original “band” that did not fall off and helped every bit they could through those years.

Special Thanks to:

DariusSheeno
CH FR
WoozyWinkle
nenza

CHFR

Drafted in the Wide army

I discovered Hidamari Sketch in early 2020 as Majormilk kept talking about it on the server. Just like him, I pretty much dropped it at the end of season 1 and didn’t touch it for another 4 months. It was only in the middle of June that I picked it back up, and after having to ask what season 2 was even called, it absolutely rocked my world and I became hooked!

x365 got so deep into my mind that it started overwriting fundamental mathematical constructs in my brain.

Jokes aside. Hidamari Sketch really did have a huge impact on me. Back in July 2020, I was going through some stressful times: the school I got accepted in was at risk of closing. I was fresh out of high school and had no idea where things would go if that door closed on me. All I could do was wait anxiously for additional news, during that time, Hidamari Sketch was my beacon of light: The relaxed atmosphere, and Miyako’s philosophy of life (don’t stress over it if you can’t do anything about it) helped me cope.

In the end, everything turned out alright. Which I’m grateful for.

Working with MJRMLK

Now, I’ll talk a bit about my experience with collabs.

Over the years, I’ve had the occasion to work in many projects and with many people. Most of the time It’s a fairly straightforward process, I do my part, submit, and voilà. I haven’t had to compromise or give much thought to the “collaboration” part of the work.

Majormilk is the one person I’ve worked with the most when it comes to Otomads, thanks to our similar tastes when it comes to anime, things usually just click. However, when working together, things suddenly become a lot more involved:

• Tons of messages are exchanged over the course of long brainstorming sessions to come up with a story for our part;
• I am made to go through many checks and “red corrections” to even get a passing grade art-wise;
• Then, I have to accommodate voice lines from Non Non Biyori to create a fictional scenario that is BOTH a crossover with Machikado Mazoku AND a reference to Dragon Quest;
• And finally, I am made to take part in a “coloring assembly line” with live tutoring to help with the many hand drawn assets of our part.

And all of that was just for one single part mind you! And what’s even worse is that I can’t even complain because Majormilk works himself even harder on the visuals. Seriously, a lot of the stuff he does is made completely from scratch!

Despite how involved those parts usually are, they’re still loads of fun, so I was very excited to take part in Super Hidamatsuri when it was announced: not only would I, once again, get to work closely with Majormilk, but it would also be on an anime that means a lot to the both of us. For that reason, I was ready to give it my absolute best.

Working on my part

Disclaimer: while I am talking (rather strongly) about the frustrations I’ve encountered making my part, I regret absolutely nothing. I am incredibly proud of having contributed to this collab and everything I went through was worth it. The following section is meant to be read as a recounting, not a rant.

The audio

To say that this part’s audio kicked my ass would be an understatement, I was at WAR with this audio! To give you a basic idea, my first preview was from March 2021, and my final submission was done in November 2023, long after I completed the visuals.

Let’s talk a bit about the medley, I’d like you to imagine yourself as an audiomaker, now imagine having to work with a medley that is made only of character songs and openings, and that the source you’re working with… is the very series with all those character songs and openings.

Because of (or rather, thanks to) Majormilk not being an audio guy, he went all-in with this Hidamari-only medley, without realizing how terrifying of a prompt “make an interesting audio where the girls are covering a song that they are the initial singers of” is to an audiomaker.

As if that wasn’t enough, my inner showoff decided to pick one of the more high-stakes opening, “Hatena de Wasshoi”, and live up to its legacy.

And so, I started working on my part.

First, I tried playing it straight by using the original lyrics. Taking the voices from the original song was impossible since the medley was five-or-so-semitones higher, so it would have sounded like trash (and my pride wouldn’t allow it).

I did the first thing that came to mind: voicebanks of the four main girls. Making a Voicebank sucks, it’s soul-draining, you spend hours looking for all the syllables, preferably with duplicates in case some of them don’t play nice with UTAU, and then, you have to configure all of said syllables. It’s very repetitive, and yet still requires your attention since all syllables need different values for their config.

So I did a voicebank… four actually…　and guess what:

It fucking sucked lmao

Sticking with the original lyrics was simply too boring, so I added some sentencing elements in the first section of the song to make it more interesting. From this point on, I also started working on the visuals in parallel as they would drive some elements of the audio. For example, I decided very early on that I wanted to pay homage to the scene where Yuno rides the truck to the Hidamari apartments in the first episode of x365.

With UTAU’s processing, the voices were blending together a bit too much, I tried making two of them sing harmonics:

I got a passing grade, but still felt like the audio was missing something at a fundamental level: it was way too boring.

In this state, my part was at best a mediocre recreation of a good opening. I decided to put my worries aside for the time being to focus on the visuals instead.

Eight months later, I went back to the audio to give it another shot. I restarted from scratch.

First, I ditched the UTAUs in favor of a more manual approach: roughly pitching in Reaper, and doing additional tuning in Vocalshifter.

I also tried to do something à la “brazilian dog dance” out of desperation, maybe something good would come out of it.

I didn’t like it and scrapped the idea

At the end of the day, I still felt like something was lacking: having the girls sing their own opening song just wouldn’t cut it. For the third time, I restarted from a blank slate, that’s when the idea hit me.

Time to put my JLPT N6 skills to use, I would change the lyrics of the opening to make them sing about creating an Otomad instead!

I was able to get valuable help from other participants.

Things were starting to look up, I was more content with the story the part was telling and did a better job with tuning the voice.

I decided that this would be good enough, and declared my audio as “done”. I was able to execute my ideas and tell the story I wanted to tell, so I started fully focusing on the visuals.

The mixing was incompetent, but I could live with it… until I couldn’t anymore.

Cue in two years later. Mikami started working on the collab’s mastering. I had piled up on experience and was ready to throw hands with my audio one final time.

What followed was an intense learning session where Mikami gave me invaluable help to improve many aspects of my audio.

From natural singing

To chords

To how OTT works in greater details

To long sessions about mixing, balancing, and making my audios less muddy.

I have learned a lot from these conversations, for which I cannot thank Mikami enough.

At long last, I had won the harsh two-year-long battle against my audio part.

The visuals

The visuals for my part were a different kind of battle. They were not draining because of a lack of skill or anything, they were draining because they took so much god damn time to make!

As I fumbled around with the audio, I started working in parallel with the truck scene.

I started by syncing Yuno’s sounds to different “obstacles” that would rock the truck around.

You can see some of the ideas I started with, they all got scraped in favor of a more stable scene that doesn’t focus on the truck movements as much. As for the last video, I believed I wouldn’t be able to translate that angle change well in AE

For the second scene, I initially wanted to use the Hidamari SVG character icons (courtesy of NoAddedMSG) with backgrounds that kinda represented them (Miyako was the sunshine in Sunshine Sketch, Hiro had food, Sae had a manuscript, but I couldn’t think of much in Yuno’s case thought).

This is the initial draft of the scene, after a bit of polish, it looked like this:

There was too much contrast between each shot, and Yuno’s icon disappeared too fast, causing an awkward rhythm. That’s why I replaced it with the room interior scene, which flowed much better, using Hoshimittsu’s production notes as a reference.

(For fucks sake affinity please make a linux Port)

And then I ditched After Effects. I basically rendered the truck scene as a PNG sequence and never touched an Adobe CC app after that.

I was learning Fusion at the same time that I was making this part.

This scene was very fun to make AND looks super good, the way in which I connect the same scenes to a different renderer for the final shot is something that’s made so much easier with nodes!

I then went on to model the hidamari apartments (while Milk was working with Darius to model the Hiro/Miyako Robots at the same time)

The scene with all the audio items was a blast to make, I got a bit more used to Fusion’s 3d and particles systems, having to mix both the openGL renderer (for custom particles), and the software renderer (for proper transparency). Nodes proved extremely useful there too.

once I had an animation for each audio item, all I needed to do was duplicated it and offset the time to get a very nice scene.

Just as a side note, can you believe that After Effects still does not have a “duplicate” effect? Be it in 2d or 3d space, AE users are forced to either mess around with expression and layer indices or use Trashcode Echospace.

I had to draw a bunch as well:

I entrusted Majormilk with my life, and as expected, got hit with many (much needed) corrections in red.

In one of the later scenes, I used a mist pass there to get a depth reveal for the colors. I had wanted to do this kind of effect for a while.

After finishing these visuals, and going back to fix the audio, I was done. I finished this part, it was one of my most intensive, but also proudest work.

A sidequest before the finale

One week before the deadline, Majormilk made the announcement that Hajimari Survival still had a blank spot visual-wise. Aya was having trouble finding ideas so he asked for anyone who wanted to fill in.

Hajimari survival was actually my second choice at the start of the collab, I liked Miyako a lot back then (then I turned into a Yunofag). So I voluntered to do those 7 seconds of visuals.

This time, I used many pre-existing assets (mostly Kirafan illustrations). Because I had to hand-craft almost nothing at all, the part was done in a flash. It only took me about two days to hand in the finished product:

This video was made with Autograph (my second-ever project with this software after IOU).

Final thoughts

I am very glad to see that this collab came to fruition, both because I believe the end product turned out fantastic, but also because it is a huge relief to not have it occupying a corner of my mind every single day. During those four years, I’ve had countless moments where, when idle, my brain would automatically go back to brainstorming audio or visual ideas. While this in itself was fairly common and happened for most of my projects, 4 years is just way too long for a hobby project to stay on your mind, it definitely wears down on your morale and motivation. This is particularly true for Majormilk, whose mental state was sad to see during the collab’s darkest hours.

I think Super Hidamatsuri is a work that is very reflective of the host’s vision. The heavy-handed management is not something that actually happens in that many collabs. I think he managed to orchestrate a cohesive visual flow without limiting each participant’s visual style.

More than anything, this collaboration is a huge passion project, born solely from the love of Hidamari Sketch. We knew from the very beginning that this collab would barely break past 10k views. We also knew that neither the source material, nor the medley, would be appealing to madmakers due to the lack of popular otomad BGMs. But despite all that, we still pushed through and were able to get it done.

From the bottom of my heart, I was glad to celebrate Hidamari Sketch with this collaboration.

Thank you. Stay wide.

CHFR.

A message from everyone

We asked in the server if anyone wanted to leave a small comment about the collaboration. Here are the responses we received: