We want to give a big thank you to everyone that watched, enjoyed, or gave feedback on our work! (っ^ ◡ ^c)❤ We're really pleased we could finally share it with you! We have a lot of exciting plans ahead and made some big changes to how we work already for Melody. We'll be returning to sharing things much quicker ٩(。•`◡´•。)۶
We definitely understand the frustrations with waiting. Over and over we found ourselves surprised at just how much a long duration affects our entire process. We made a lot of estimates based on our old shorter projects and between the estimates we shared, and the ones we kept to ourselves, all of them were just completely off. We found that after a certain point double the duration isn't double the work. It's probably triple. And Fischl being 6x longer than our usual stuff really pushed us in ways we never anticipated.
But regardless, we apologize again for the overly optimistic estimates. We were just doing what we had always done for the projects before, but they were all short enough that it never bit us. We learned a lot from this one and will be more careful about estimates in the future because of it (T ^ T )ゞ But we appreciate how patient many of you were and look forward to pushing for a better release schedule for you from here on!
Beyond that, there's a pretty mindblowing amount of made up claims or deliberate misunderstanding about many things regarding this project. "$10,000 microphone", "2.5 years", "fixing lights 10,000 times", "repeating the same scenes over and over", "4 months of retakes", and some others I probably missed.
I have no idea where the $10,000 microphone thing came from, maybe it was a joke somewhere that was misunderstood. Or just a goofy lie for some reason. But no, we first bought a cheap mic for under 200 that another voice actress we worked with used. Over a year or so we rolled with it, tried some software tricks like equalizing or other tools, more sound deadening for the environment, etc, and the mic just sucked no matter what we tried for the WIPs. So we bought another one for about 350, that also came with a decent interface. Massive boost in quality even without software, you can easily hear it for yourself by comparing early WIPs to later ones.
That new interface also died less than a year later sadly, so bought another one for I think 150. It changed the sound profile a bit (hard to hear in standalone WIPs, but noticeable when scenes are strung together back to back). The mic itself was still fine though, it's been working great for a while. But we had 3-4 varying quality and sound profiles between hardware and settings during those WIPs, so obviously we couldn't use a lot of it for the final. It doesn't match, and it's obvious. Compare it for yourself and you'll get it right away. The voice had to be redone so we just bit the bullet and did what we had to do. Nowhere even in the stratosphere of $10,000.
Even if the quality weren't so inconsistent due to hardware and software, voice actors aren't machines. Only a fraction of the project had voice, and no voice actress we've ever worked with can nail the perfect matching sound and emotion on different days, let alone the first try, to fill in all of those spaces. So we redid the whole thing, using the good parts from WIPs as references, so that it would actually match better and not sound like thousands of moans got scrambled up in a train crash.
For over a year we've heard stuff like "it's been 2 years" in comments, especially Twitter. But it hasn't been 2 years yet, or "2.5 years" for that matter. We're about a month away from the 2 year anniversary still. In fact, we only just passed the 2 year anniversary for Ei x Miko. In the last 2 years we:
- finished and released Ganyu's solo animation
- made Succubus Hu Tao
- did a couple of small one off animations (Ganyu x Keqing, Yoimiya, Mona x Fischl x Slime)
- tons of wallpapers
- got a solid start on Melody
- and made Fischl x Slime
23 months, for all of those things. Not just Fischl x Slime. Adding the several minutes of other animation work, and weeks of time experimenting with some wallpapers for fun and preparing additional characters models for them, we're surprisingly still extremely close to our usual 1.5 min/month pace. Mona x Slime is a bit of an outlier, since as mentioned it had a lot more repetition. We put far more into facial expressions and custom motions for Fischl than for Mona. Fischl has far more scenes of custom motion instead of just loops. And Fischl got proper voice acting, where Mona was just scrambling a small library of voice clips repeatedly. I'll never say your preferences for one voice over another are wrong, but on a technical level Mona's was far more simplistic, repetitive, and not tailor made for each scene. And similar for the animation, Mona was just more simplistic and repetitive at every level. Not out of laziness, just that was the best we could accomplish at the time. But that also made her animation one of our fastest paces ever, especially for longer works. We actively try to avoid some of that now to focus on better quality so it's hard to say if we could ever hit a pace like that again.
As for the "fixing lights 10,000 times" and "contradicting ourselves in every post", there's really 2 aspects to it. One, we could have done a better job communicating some of those things. We got a lot of feedback there and figured out where some of communication breakdowns were happening so we'll do our best to improve on that front with some plans we've made. And two, some of what we said was pretty clearly just misunderstood or exaggerated. Stuff like the "phases", and going from a later phase to an earlier phase, and "polishing".
Contrary to the claims we "never did that for previous animations" I saw here that's actually pretty far off the mark. Every animation needed polishing, but in older works we didn't go into so much detail and try to discuss every little thing in the process. Mostly because they were so short that there were far fewer WIPs and updates. We shared a ton of information for Fischl like never before, but didn't realize we needed to share even more to give it some proper context. Every animation had core lights done for scenes, then checked again later, updated and finalized after animation was complete. Every animation had core motions done first, then time spent later polishing and adding more to it after the full core motion was done. It's part of the process. Always was, always will be. But we have some plans to better describe these things, and give more context to future updates.
The "repetition" mentioned in scenes is only slightly true. For Mona, yes there was quite a bit that could be repeated with minimal work. That made Mona abnormally fast to produce for that length. Not for Fischl. We did use a few of the same or very similar camera angles. Some parts were so long we literally ran out of good angles and didn't want to settle for bad angles just to make it look different. But in spite of similar or same camera angles, for only a handful of scenes, every scene is animated from the ground up. It's not the same motions on repeat and that should be pretty clear from changes in pose, speed, or especially facial expressions and custom body movements and reactions. Even if it seems the same to you, it required just as much work as any other scene except maybe re-using lights for a couple scenes with tiny adjustments.
Fischl has more scenes with dynamic non-looped motion than probably every other animation we have combined. Loops save a bit of time and effort but we enjoy the story stuff a lot and went even heavier on it for Fischl. But even with the looped stuff we also made sure that, unlike Mona with just the 1 pose really, we constantly put Fischl through changes in the action. Doggy style, face down, upsidedown, crawling across the room, on her side, on her back, just all kinds of dynamic shifts in poses and action. And lots of little reactions or expressions changing even during those looped parts. Mona had far less of all of that so they're really not comparable.
And the "4 months of retakes" thing. I double checked our chats with the voice actress, and we first sent the video for the very first takes for the nearly finished animation in early March. It was almost exactly 3 months to prepare the entirety of the audio from zero to release day. Recordings, retakes, voice direction, sfx, ambience, music, reviews, polishing, sourcing new sounds, and a couple weeks we lost by trying out a new sound library that just didn't work for us and had to be redone in some parts. Everything else just the usual create > review > polish process that applies to all of our work. So no, it wasn't 4 months of retakes, it was 3 months of the entirety of the audio work, and during the first month of it there were still some last bits of animation polishing, re-renders, etc on the side.
Finally the lack of WIPs at the end and "proving we're doing the work". We have a discord server where we've streamed probably more than 5000 hours of our work the last couple years. We practically live there. We more than proved we're doing the work there day after day, again and again. Now to be fair we shouldn't have paused the WIPs, but that's easy to say in hindsight. We expected a couple weeks of work, so just started crunching to get it done without the extra overhead and time eaten up by preparing WIPs. Obviously that didn't go as planned, but you live and learn. You can pretend we just had a 3 month vacation and did nothing, deliberately risking our entire image and fanbase and future in this line of work. But it's simply ignorant, and the logic there makes no sense. The faster we release, the more we can hype up the next project, the more people we can pull in, and the more we grow. "Stalling", "milking", or whatever else you've imagined is beyond counter productive unless we were planning to quit for good. And we aren't, we've already had significant progress on the next animation the last couple months.
But sandwiched between some of these absurd claims and insulting remarks there were some fair bits of feedback we'll take into account though. We appreciate the sincere insights into the parts you did or didn't like. As always we're continually looking to improve and trying out new things as we go so we hope it will make future projects even better! (and much quicker)
Hope this helps clarify some things. We're also planning to give a better breakdown of whole process of creating our animations so that it will be easier to keep up with progress updates and see how much is left. We may post some version of that on Twitter as well so it's not just paywalled, will see how the plans go. And once again, thanks for watching! ❤(˵^ ◡ ^˵)❤