Thread by @Lystrialle

This will probably be another long thread, but since CC2 is finally done and I probably have a new Vocaloid release on the horizon, I kind of want to talk about the difference between how I approach ARM work vs. how I approach Vocaloid work

I think there's really no denying it that any work I make for ARM, musically, is probably significantly higher in complexity, and possibly overall better quality than anything I make with Vocaloid at an equivalent point in time

So the most obvious potential reason is that Vocaloid's hard to work with, so I won't make complex work with it, but that's actually not the case at all -- working with vocalists isn't as "easy" as it sounds (you still have to do vocal re-centering and post-processing and --

-- timing, whereas I'm so used to the Vocaloid workflow that I can probably get a tuning job done in 1-2 days, UTAU probably in 5-6)

So the real reasons boil down to 1) workflow and 2) timing.

Re: workflow, it's important to note that my Vocaloid work is fundamentally my "solo work", and my main reason for using it at all (besides "because I like it") is that I can do everything top-down

it's really the one place I get to be control freakish if I really want to be, so I do everything from composition to arrangement to lyrics to mixing myself, even when I'm not as confident in it

With ARM, I only do composition and arrangement, and sometimes even not by myself

All the songs I did for ARM had at least four people working on each one, so to be quite honest I think it's pretty unfair to compare a work made by multiple people to a completely top-down onemanned one, regardless of what kind of singer you have...

Re: timing, a really important thing to note is that the song I just put out on CC2 side red? Yeah, that was in development hell for two years. (I'm comfortable revealing this because we released a very different-sounding demo at Winter Comiket 2016.)

ARM is an event circle, so releases have a *very* long span of time between them, and we really need all that time because we're checking lore compliance and coordination between multiple participants at the same time. That amount of time is totally impermissible for solo work.

Other than compilation projects, all my solo deadlines are self-imposed, and I deliberately have to stop myself before I get too perfectionist, otherwise I am not going to get *anything* done in a timely fashion. That reflects badly on both me and my own skills, too...

I'm pretty self-conscious about the fact I'm the youngest and generally most inexperienced on the Cosmocrystal composition team, so I really don't want the gap in quality to get too large. The two years of development hell I mentioned were full of me trying to improve my skill.

It's like this: xkcd.com/1414/

Spending time doing *more* projects, even if they're done quicker and less refined, often ends up developing you far more than doing only one and being perfectionist about it. Sometimes there's a point you really have to let it go.

If I had been working only on year-long projects between 2016 and now, and figuring out things about composition, there's a genuine possibility Quinton's song would have stayed unfinished 💦

And of course it reflects backwards, too. Aria's song in 2017 made me really rethink --

-- my approach to arrangement, so I think it definitely had a very large impact on my solo work thereafter. Because I had that opportunity to really challenge myself on a project like that, it exposed me to a lot of experience I may not have had otherwise.

Anyway, that's how it goes. I think working solo and collaboratively, with Vocaloid and without, has allowed me to be more well-rounded and become better at both types of work. So although it's tough to balance both, I hope it makes better results for you guys to hear, too 😅