
BY Yiban Youqu (一般有趣).
1. Introduction
Lab on Roof × ZERO MAD TEAM.
Through this festival, Lab on Roof and ZERO MAD TEAM hope to provide a welcoming forum for still-image MAD—fan edits made by animating and compositing still images—and opportunities for creators to collaborate and discuss their work, while also offering viewers a distinctly Lab on Roof approach to appreciating the form.
2. Closing Address
Working closely with ZERO MAD TEAM, and with the support of independent judges including Zhizui (支醉), Luci (鸬鹚) and Fengyuejiang (风月酱), Lab on Roof has now completed the judging for this Still-Image MAD Festival.
Honestly, this was our first time organizing anything like this. With no relevant experience, I was pretty anxious, and yes, we did keep people waiting and run into setbacks. But thanks to the kindness and help of the creators and everyone involved, I think, as the organizer, that the event as a whole had quite a lot going for it~
Some of you may be wondering why Lab on Roof organized this event. The answer is really quite simple: our interest in and love for still-image MAD as a moving-image form, and for its possibilities. As for everything else we feel, I think I’d rather let the works in this festival express it than put it into words. It’s all there in them—
Here are the eligible entries in this festival, in the order they were entered on the record sheet. Danmaku are on-screen bullet comments.
Now, let’s announce two rather special awards first, just to liven things up!!
ZERO MAD TEAM Leader’s Choice Award
Da-da-dum (pretend there are drums…)—!!
【Wu Xie-Centered/Still-Image MAD/Sha Hai】The World Is Full of Things Beyond Understanding. “0305 Happy Birthday, Wu Xie” (03:10; figures shown on the page: 37,000 views, 422 danmaku)
Lab on Roof Director’s Choice Award
Boom-boom-boom (pretend there are drums…)—!!
【Still-Image MAD/Hayate the Combat Butler: Royal Castle Arc/Golden Festival 2019】A Wish Fulfilled After Ten Years Apart (03:20; figures shown on the page: 7,974 views, 20 danmaku)
Tempted to click over and watch? If you are, please do—really, it’s fine!
There are more big awards coming later, but let’s keep you in suspense for now! Next up are the critiques. Let’s use them as a way into what makes still-image MAD, including the winning works, so appealing and interesting!
3. Comments on the Works
【Still-Image】I Want to Eat Your Pancreas
Runtime: 02:18; figures shown on the page: 4,224 views, 12 danmaku.
Zhizui:
- I didn’t understand what the white line in the background at 0:37 meant.
- At 0:45, I’d suggest moving right first, then left. There’s text, after all: moving from right to left while the text has to be read from left to right creates a bit of a conflict.
- At 1:13, is that a tear falling and then dispersing? It isn’t clear. There’s also the line “a sudden death” afterward, and the whole image lacks a clear visual focus.
- I think the composition of the final “goodbye” could still be improved. Red, white and black currently take up rather similar proportions of the image.
- There’s quite a lot to praise, so I’ve just offered some suggestions~
Fengyuejiang:
- The overall level of finish is decent. Some details could be strengthened.
Luci:
- There were plenty of comments in the newcomer competition already. Personally, I think the bigger problem is a slight lack of variety and standout moments. Try not relying on text next time.
Lab on Roof:
- The black, white and red palette feels really good with the instrumental music. Around the vocal pause at 0:30–0:35, I feel the transitions into and out of it, and the fit between sound and image, could be better~ In terms of emotional expression, the visual narrative reveals Sakura’s fate midway through, then moves into “my” remembrance and farewell. Bringing back the diary, Living with Dying, as a narrative thread in the later section would give it more resonance~
Video status error (av86527602)
Zhizui:
- Don’t let the movement stop. The last few frames of a lot of shots freeze.
- The style isn’t consistent enough, and sometimes the tone suddenly gets brighter. Why is the section at 0:56 so bright?
- Some of the 3D is unnecessary. Take “defective,” for example: I had to pause for a few seconds before I could make it out. Leaving the rendering quality aside, surely the lettering should at least look “defective”? But the characters are perfectly intact; they don’t look defective at all. You could add some fragments and debris around them, too.
- You need to use particles and noise more artfully. It’s particularly obvious around 0:40–0:50. It’s not enough just to have lots of them, and it’s not enough just to keep them drifting.
Fengyuejiang:
- The 3D compositing and the motion curves in the shot transitions are poor. It all feels like a bit of swaying and shaking.
- Have a look at Oomugi’s (大麦) MADs. There’s a lot you can learn.
Luci:
- It doesn’t come together as a whole; it feels very much like an exercise. Keep at it.
Lab on Roof:
- The font choice feels quite right for the story, but English seems to appear in some places and not others, and I don’t really understand the thinking behind that. At 1:30, the distressed-image effect used for the girl on her journey resembles the one in the earlier memories, and seems at odds with the brighter feelings on either side. I like the way the ticket is presented at 1:10. The music seems to fade out a little too hastily at the end, falling out of step with the images.
【2020 MAD New Year Festival Individual Entry/Still-Image】Arknights [War · Purge]
Runtime: 03:07; figures shown on the page: 5,162 views, 10 danmaku.
Zhizui:
- You could lengthen the transition at 0:18 with Motion Tile or something similar. Don’t just switch between the two scenes: generally, you want to add one to five completely blurred intermediate images that don’t belong to either scene. The transition’s duration needs to match the music.
- Be bolder with the lighting; don’t be afraid of overdoing it. Add more of that divine light streaming down through the clouds, that Tyndall-effect sort of thing. Each shot lingers a little long at the moment, so richer content would help. But by “stronger,” I don’t mean quantity—I mean intensity, in brightness, size and so on.
- The rain at 1:20 is too timid as well. I had to look closely for two seconds before I could tell there really was rain.
- For the blinks at 1:40 and 2:37, unless there’s a particular expressive purpose, there’s no need to blink twice in such a short time. If you insist on it, give the character a different expression before and after—more of a rueful smile afterward, for example—instead of just blinking for the sake of blinking.
- There aren’t quite enough elements built up in the final scene. You could make it more epic.
Fengyuejiang:
- The compositing is quite good, and so is the design, but you need to study transitions.
Luci:
- The visuals are clean, but the staging stays too close to the source material to carry the music. Not enough rise and fall.
Lab on Roof:
- Within individual shots, I can feel the camera movement directing the eye toward the subject of the narrative. Several of the hand close-ups are very good. I think using a wider range of shot sizes in combination would make it even better. Try more variety in the transitions and changes in frame proportions next time, too~
【Still-Image MAD】The Emperor’s Academy
Runtime: 01:12; figures shown on the page: 1,805 views, 1 danmaku.
Zhizui:
- Give the font choices a bit more thought.
- Big blocks of text are hard to read. Some still-image MADs use huge blocks deliberately made illegible, as visual elements. Here, though, you actually want people to read the text. The wording needs work.
- The backgrounds are a bit monotonous. Avoid conspicuous edges, especially the kind at 0:07.
- Remember to align things with the edge. At 0:11, for example, there’s a white border beside the character on the left.
- You’ve still got some learning to do with text composition, kid. A lot of the text flashes away before my eyes can even move over and make it out.
- It’s best not to stop the camera movement. Keep things moving.
- The text flying past at 0:15 doesn’t show who’s speaking. Some dialogue needs to be composed together with the character; otherwise people will generally read it as narration.
- The light at 0:31 isn’t plausible. That light spot couldn’t be elliptical like that. Don’t just stretch it by scaling. I won’t even get started on the 3D.
- Overall, the backgrounds are rather monotonous, and it isn’t very fluid either. There’s a lot of room to improve. Also, remember not to make the eyes blink so slowly. You could film yourself and see how many seconds a blink takes.
- At 0:48, I suspect you took a shortcut. It’s not unheard of to reuse one large image, but just moving sideways across it looks cheap. Blur the surroundings, add flickering, add glow, rotate as you push in or pull back—whatever you do, it can’t be so plain.
- At 0:50, the ripples are going over the shoes. Try stepping in water yourself and see whether the ripples go over your shoes.
- At 0:53, don’t go straight to a white screen. You could have a huge sword flash sweep across the screen before it goes white.
- You could save that final GIF for later, as a way to gauge your technical progress—come back and make it in six months or a year, for example.
Fengyuejiang:
- The overall level of finish is low, and the motion curves are rather stiff. Spend more time looking at how the original video does it.
Luci:
- Quite bold, but you still need to work on the visual execution.
Lab on Roof:
- The editing pace feels somewhat out of balance both within the visual narrative and in its relationship to the music. You could try adjusting the frame proportions and shot size within a single shot to invest the subject of the narrative with the feeling you want to express, rather than rushing to cut or moving too quickly.
【Still-Image MAD】Her Wings
Runtime: 03:47; figures shown on the page: 8,485 views, 35 danmaku.
Zhizui:
- Very polished. You could easily get viewers interested in the source material with this. I’ll skip the good points~
- Personally, I find what’s at 0:32 and 0:41 a little too thick. It doesn’t look good.
- The line of text at 1:08 doesn’t have to disappear immediately. It goes too fast right now: I’d only read the first half before it was gone. You could actually keep it there and fade it out later; it would still fit the rhythm.
- The hair movement at 1:36 could be improved. The shot is short, so it looks all right as it stands.
- At 1:48, does something go wrong with the mast for one frame? Why does a black stick suddenly flash up?
- At 1:59, I’d still recommend not squeezing the image. Find a way to add something on either side. The distorted proportions do affect the viewing experience a little.
- The background construction at 3:31… doesn’t look very good. It took me out of it. _(:з」∠)_
Fengyuejiang:
- I feel the overall arrangement could still be improved. The way the 3D sections blend into the whole feels a little odd; I hope you can improve that.
Luci:
- The opening goes hard, but what follows can’t quite sustain it. Vary the emphasis according to where the weight of the piece lies.
Lab on Roof:
- It uses the full song, with the visual narrative working in harmony with it, easing off at the right moments and telling the whole story. At 1:35, the close-up of 千奈 → full-body shot → long shot plus foreground effects—I really like that editing~ The shot at 1:58 doesn’t quite connect with the previous one in how it guides the eye. It took me a good while to spot 千奈 right over at the edge on the right. ( ´Д`)y
【Valentine’s Day Against-the-Tide Vanguard · Still-Image MAD】The World and a Committed Singleton’s Travels: Dammit, Shoo—It’s the World I Love
Runtime: 01:36; figures shown on the page: 4,759 views, 3 danmaku.
Zhizui:
- The circles in the opening are too thick. And actually, white circles would work just fine.
- The butterfly stops a bit too abruptly at 0:23. Adjust the motion curve some more; make the stop gentler.
- The fonts are a little inconsistent. You don’t have to use just one, but they still need to share a stylistic direction.
- I’d suggest making 0:28–0:34 feel less high-tech. Suddenly throwing that into jazz hip-hop is a bit jarring, unless the whole piece has that high-tech style. The main direction here should still be poetry and faraway places.
- The idea at 0:52 is very good, but adding smoke footage between the images to ease the transition, with another layer of movement, would make it even better.
- I think the section at 0:59 could still be improved—made more refined, and more poetic.
- The composition at 1:13 is excellent.
- Also, why is the prologue’s title in Chinese? Surely the titles should at least be in the same language. _(:з」∠)_
Fengyuejiang:
- The design could be stronger. For this, my first recommendation is Oomugi’s still-image MADs. There’s a lot to learn.
Luci:
- A fairly freewheeling exercise.
Lab on Roof:
- I really like the prologue’s panels and the fluid transitions as things enter and leave the frame~ The editing at 1:10 feels a little rushed. How should I put it? Quite a few ideas in the later section seem to pass by too quickly. The tenderness and sadness of wind, clouds and night are reduced by the brisk cutting and music to much the same fleeting impressions. But overall, it’s still a good watch~
Video status error (av86077404)
Zhizui:
- It is a collaboration, after all, so I won’t get into the radically different styles and the visual gaps where they don’t connect.
- At 0:16, some lighting on Aoba would help bring out the sense of space.
- There’s animated footage you could use for the water at 0:23. Have a look around; it would work much better.
- Why are frames mysteriously dropping during the final pullback at 0:26?
- You could do more with the image in the shot at 0:44. At first I thought you’d lazily used a screenshot; it took me two seconds to realize it was that picture from the show. Even so, the shot is still very basic.
- The three moves at 0:55–0:57 could be faster. Adjust the curves some more; the first two could be shorter still.
- The scaling when it pops out at 0:59 might distort it just a little too much.
- That sphere at 1:10… leaving the rendering aside, adjust the color so it fits. Just sitting there like that, it looks a little out of place.
- Finally, my favorite, Yun, only gets to have her name shown in the group photo… _(:з」∠)_
Fengyuejiang:
- Quite polished, with good motion curves and pretty good color grading. Pretty good motion graphics for moe pigs—cute-character obsessives—overall.
Luci:
- Not bad. It could be more fun.
Lab on Roof:
- Bright music and colors deliver a dose of cuteness. The moon-to-sun match transition at 0:20 is quite good. The wave effects could use some more adjustment. At 1:08, the shot starts with a close-up of the lettering; the lettering and the Earth then obscure most of the main character imagery. It’s a bit baffling.
【Still-Image MAD】What I Thought Was Unrequited Love
Runtime: 01:49; figures shown on the page: 5,870 views, 4 danmaku.
Zhizui:
- Fluid and really satisfying. I particularly like 0:40–0:48. Basically, I couldn’t find much wrong with it overall. If I had to say something, take another look at the train’s shaking at 1:37. The vibration of a moving train isn’t just straight
wiggle; pay attention to how trains shake when you watch anime from now on. - Friendly reminder:
wiggletakes five parameters inside those parentheses, not just two~ You could also try an expression using trigonometric functions for Position.
Fengyuejiang:
- Good motion curves, but there are basically no background elements, which leaves the whole video looking rather flat.
Luci:
- Find a way to handle the less prominent sections better and give them more to hold the viewer’s interest.
Lab on Roof:
- You’ve deliberately used the manga panels, and even the speech bubbles, to create close-ups that naturally express the characters’ feelings. Really nicely done~ The cuts work well with the music, too. The presentation around 1:20–1:30 might be worth another look; the effects and the switches between them feel a little too frequent.
【Still-Image/Confess Your Love Committee】The “I Like You” on the Tip of My Tongue
Runtime: 01:47; figures shown on the page: 4,048 views, 3 danmaku.
Zhizui:
- The cut at 0:15 is a few frames late; it misses the music.
- At 0:26, the image could do more to express what the text says. Maybe through the characters’ relative sizes in the composition, differences in color brightness, differences in blur, or a hook made of text between them. Whatever the means, make the image convey being so close yet so far away.
- At 0:27, please don’t suddenly slip Japanese in there, haha.
- Why does the pullback at 0:40 drop frames? Did you remove some?
- The thing at 1:10 takes me out of it a bit. It might be better just on the face, not on the hair.
- At 1:20, you could use Puppet to give the fist a little movement and show that resolve.
- The scene at 1:30 doesn’t quite fit, but it’s not a major problem.
- Overall, excellent.
Fengyuejiang:
- The level of finish is already very high. Some details could be strengthened.
Luci:
- For this kind of piece, the shots carrying the most weight could be cleaner.
Lab on Roof:
- The treatment of color, contrast and so on fits HoneyWorks’ atmosphere of youthful romance quite well. The odd thing is that in some places the actual movement seems to stutter? The one-way arrow at 0:57 and the puzzle that follows express Kotaro’s feelings for Hina. The design itself is very good, but the composition and editing make those feelings a little hard to grasp. You could find a way to bring them out more. The overall level of finish is very good~
【Still-Image MAD】Last Train【BEASTARS】
Runtime: 03:36; figures shown on the page: 3,704 views, 4 danmaku.
Zhizui:
- Very understated, and really pleasant to watch. It just tells the story straightforwardly, without any flashy stuff.
Fengyuejiang:
- The shot arrangement is quite strong. It’s already very good; you could also watch av11986051 and learn from it.
Luci:
- Next time, try putting more emphasis on atmosphere with this kind of piece. The form is a little thin at the moment.
Lab on Roof:
- The skillful use of panels, frame proportions and composition—the selection of images from the manga—brings out the characters’ feelings and the atmosphere of the scenes very well in the visual narrative. The song “Tokyo” fits the story’s atmosphere beautifully, and the editing keeps pace, sensitively conveying the pretty bunny girl’s delicate inner feelings. Love it, love it.
【Golden Chicken/Masamune-kun’s Revenge/Still-Image】Masamune Really Is a Man You Love and Hate.
Runtime: 02:52; figures shown on the page: 4,983 views, 13 danmaku.
Zhizui:
- Why does it feel as if frames are dropping here and there throughout?
- I wouldn’t recommend the pure black and white you use at 0:45 and 0:52. It makes things a little inconsistent. Actually, the scene construction could use a more consistent style too. And I don’t know what that scene at 2:08 is for; it feels a little unnecessary.
- Oh, and the English subtitles at 0:48—you forgot to change the text after duplicating the layer.
Fengyuejiang:
- The level of finish is quite good. The details in a few shots could be strengthened.
Luci:
- The visual execution shows solid fundamentals, but personally I find this pairing of material and music neither here nor there. Combined with the frame-skipping feel, the contrast is a little strange.
Lab on Roof:
- The 3D at 0:20 is nicely done, hahaha. I like the way the eye close-up at 0:36 works. The visuals are highly polished, but I feel something is still missing. Thinking it over, perhaps the difficulty lies in trying to pick out the original story’s “major turning points” and use them to tell it. The original is very long, and getting all the way to the HAPPY END in the time available in the second half really is a production challenge. Rather than follow the plot linearly, you might let the characters’ inner feelings guide the telling.
【Still-Image MAD/Hayate the Combat Butler: Royal Castle Arc/Golden Festival 2019】A Wish Fulfilled After Ten Years Apart
Runtime: 03:20; figures shown on the page: 7,974 views, 20 danmaku.
Zhizui:
- Absolutely excellent. It completely blew me away. As a casual outsider, my impression of Hayate the Combat Butler was just that it had one of Rie Kugimiya’s four moe heroines—just a harem anime, right? I never imagined it had a storyline like this (though I hear this arc isn’t in the anime anyway).
- At 0:25, you don’t need to pull back so quickly. The emotion is still building.
- At 2:40, you could add a floor, send a little fine debris flying, then add some speed lines.
- I won’t get into the other little details. It’s a good watch overall, especially the chorus.
Luci:
- You’re getting better and better. Keep working on the rhythm of the shots. Keep it up.
Fengyuejiang:
- The motion curves are decent, but the level of finish is low.
Lab on Roof:
- Good awareness of composition and camera movement. The choices of color, light sources and lighting for the different emotional tones are thoughtful, too. The close-ups and dynamic action work very well with the music to stir the audience’s feelings, bringing the story fully to life.
【Still-Image】I Want You to Hear the Fireworks
Runtime: 01:37; figures shown on the page: 8,200 views, 21 danmaku.
Zhizui:
- This is a fucking beginner? You’ve got to be kidding me. (╯‵□′)╯︵┻━┻
- I can give you two suggestions at most. The car at 0:03 could bounce very slightly up and down.
- At 1:26, you could look into
wigglefor the car’s shaking. It actually has five parameters, not two. When you watch anime from now on, pay attention to how other people make cars shake. Also, you could add headlights. To get even more detailed, you could add backlighting from the fireworks on the car and streetlights. But it’s a short shot, so these aren’t serious issues.
Fengyuejiang:
- The level of finish is quite high, but it would be best to give the whole thing a consistent style. It looks rather mixed at the moment.
Luci:
- There’s a bit of a problem with how the chorus shots fit the song. Watch when you hold back and when you let loose.
Lab on Roof:
- No way you’re a beginner! I’ll skip the praise. I’m not quite clear what the light source at 1:03 is doing. At 1:05, I feel the shot of the president could follow the secretary’s previous shot simply by having him ride into frame. Moving across and adding a transition in which the foreground blocks the view seems to weaken the crispness of this moment a little. There’s also a flaw that can’t really be helped: in the visual narrative, 1:10 is itself a downward emotional turn, setting up the president’s reversal that follows, but the song stays on a high throughout. “Dramaturgy” is a suitable song overall, but here the sound–image relationship shows a crack. This bit is more or less unavoidable…
【Wu Xie-Centered/Still-Image MAD/Sha Hai】The World Is Full of Things Beyond Understanding. “0305 Happy Birthday, Wu Xie”
Runtime: 03:10; figures shown on the page: 37,000 views, 422 danmaku.
Zhizui:
- Mixing font weights is one thing, but why are you mixing simplified and traditional characters as well? _(:з」∠)_
- There are quite a lot of text-based shots. You could put more work into constructing the scenes.
- Also, perhaps a shortage of source material limited how much room you had to work with.
Fengyuejiang:
- You could put some more design work into the text.
Luci:
- Everything is pretty good. Next time, try inventing a signature finishing move.
Lab on Roof:
- I like the use of the character’s point of view at 1:05. The close-ups, the framing of parts of the body and the camera shake are handled skillfully, bringing out the on-screen character’s feelings. The palette leans dark and gray, which is good in itself, but some places are too dark, with too little contrast, and get a little muddy.
【Still-Image MAD/Made in Abyss】To the Far Side of the Abyss
Runtime: 01:33; figures shown on the page: 317,000 views, 214 danmaku.
Zhizui:
- The hair at 0:40 could use some more work. It’s moving too fast; you could lower the frequency and increase the range.
- At 0:26, the descent could continue instead of having the character stop moving. You could also add a little side-to-side sway.
- For the melting tree at 0:32, I’d suggest using a color key to isolate some orange-yellow areas, blurring them, then adding glow and overexposure. As it stands, it’s not very clear that the tree has been melted through. Did you leave the floating spark particles’ blending mode unchanged?
- At 0:35, Ozen’s left side—her back—could be darker. The stronger the contrast, the better. The shadow could enter from outside the frame, and move faster. The light and shadow cast on the wall need changes in brightness.
- At 0:50, never mind the hammer having no reflections: in a shot this long, it ought to block the sunlight, making the light flicker a few times during the push-in. The grass is too uniform in color, with no clear distinction between near and far. You could work on the tonal values here, or use blur. You could also build the grass in layers; slight color differences would make it even better. On first viewing, it took me a second to realize that was grass. It isn’t recognizable enough. If you feel there’s nothing else you can do, add a 180-degree rotation to the pullback at the start of the shot. Even that would be better than such a long shot with nothing happening.
- The abyss at the end is stunning, though it might be better if a couple of characters in some of those lines of text were red. Two other shots I like are the descent at 0:25 and the scene at 0:58.
Fengyuejiang:
- Adding some background to the opening text could actually help it blend better with the style of what follows.
Luci:
- Everything is in place. The arrangement could be more ingenious.
Lab on Roof:
- The sound–image coordination is very accomplished, though I do slightly suspect that the visual narrative is being pulled along by the song’s rhythm. The music coming in a little early at 1:16 is a small flaw. I really like the handling of light and dark, and the light sources, at 0:28. The shot at 0:45–0:50 stays still a little too long; I feel you could push in a little. Really satisfying overall.
Still-Image MAD MOON: Strange Forms on a Mad Moonlit Night
Runtime: 01:02; figures shown on the page: 26,000 views, 17 danmaku.
Zhizui:
- The first half is excellent. I suspect you’re coasting in the second.
- But the details of the movement really are very good.
Fengyuejiang:
- The design style could be strengthened. Some parts are decent.
Luci:
- Quite beautiful. Try putting more content into it next time.
Lab on Roof:
- The creator has put thought into the language of visual design. The composition of lines and shapes, and the colors, show considerable aesthetic awareness. The weakness of this piece may be that its ideas lack substance. It has a poetic feel, but still lacks a bit of direction.
Video status error (av86464329)
Zhizui:
- The shot design is quite interesting. It would be good if some shots could keep moving—push in, pull back or move sideways; a gentle movement is enough. Right now, a lot of shots come to a stop at some point. Don’t stop.
- The colors in the chorus section go a bit overboard. Or let me put it this way: watching it, I can’t find a visual focus. If I try to read the text, I can’t make it out; if I try to look at the picture, I can’t make that out either. For a moment, the viewer’s eyes don’t know what to look at.
- To be honest, I couldn’t tell what the shot at 1:12 was. I couldn’t tell even after pausing. _(:з」∠)_
- At 1:16, let the red spider lilies sway a little, then give the character’s eyes and mouth a little movement, opening wide in surprise. It would be even better if the handgun moved into readiness to fire, then a bullet took us straight into the final shot.
- Overall, the first half is good, though a few moments linger too long. I suspect some shortcuts in the second half.
Fengyuejiang:
- The overall level of finish isn’t very high. For manga MADs, you could use av11986051 as a reference and learn from it.
Luci:
- Not bad as a start. Get more practice with different ways of presenting things.
Lab on Roof:
- The typography and the shots’ color design in the later section evoke a witch’s barrier, which is good, but don’t the two halves still feel a little disconnected? You’ve noticed the religious elements and expressed them through visual language—nice. The pan across the manga at 1:03–1:09 is a little uncomfortable: the pauses and movements are somewhat abrupt. More careful treatment of the manga images would help, too.
And those are the independent judges’ and Lab on Roof’s comments on the works in this festival~
If they can help still-image MAD creators exchange ideas and learn from one another, that would be wonderful. Once again, thank you to all the judges for finding time in your busy schedules to write these comments—with some shameless pestering from me, the organizer~
Now, it’s time to reveal more awards!
4. Award Announcements
First, in accordance with the judging rules, let me announce the four works with the highest individual scores in the four categories: Thought and Poetic Quality, Visual Aesthetics, Sound–Image Integration, and Creativity in Effects—
(Ding-ding-ding—pretend there are bells.)
Highest Score in Thought and Poetic Quality
【Still-Image MAD】Last Train【BEASTARS】 (03:36; figures shown on the page: 3,704 views, 4 danmaku)
Highest Score in Visual Aesthetics
Still-Image MAD MOON: Strange Forms on a Mad Moonlit Night (01:02; figures shown on the page: 26,000 views, 17 danmaku)
Highest Score in Sound–Image Integration
【Still-Image MAD】What I Thought Was Unrequited Love (01:49; figures shown on the page: 5,870 views, 4 danmaku)
Highest Score in Creativity in Effects
【Still-Image】I Want You to Hear the Fireworks (01:37; figures shown on the page: 8,200 views, 21 danmaku)
Sadly, as stated in the event announcement, we can only do so much on fan enthusiasm alone, so there are no separate tangible prizes for these category leaders QAQ. We hope we’ll have a chance to make up for that at the next event…
—And now, the overall First, Second and Third Prizes, and the Most Popular Award!!
(Ding-ding-ding, da-da-dum, boom-boom-boom…)
Most Popular Award
- 【Still-Image MAD/Made in Abyss】To the Far Side of the Abyss (01:33; figures shown on the page: 317,000 views, 214 danmaku)
Third Prize
- 【Still-Image/Confess Your Love Committee】The “I Like You” on the Tip of My Tongue (01:47; figures shown on the page: 4,048 views, 3 danmaku)
- 【Still-Image】I Want You to Hear the Fireworks (01:37; figures shown on the page: 8,200 views, 21 danmaku)
- 【Wu Xie-Centered/Still-Image MAD/Sha Hai】The World Is Full of Things Beyond Understanding. “0305 Happy Birthday, Wu Xie” (03:10; figures shown on the page: 37,000 views, 422 danmaku)
Second Prize
- 【Still-Image MAD】Last Train【BEASTARS】 (03:36; figures shown on the page: 3,704 views, 4 danmaku)
- 【Still-Image MAD/Made in Abyss】To the Far Side of the Abyss (01:33; figures shown on the page: 317,000 views, 214 danmaku)
First Prize
- 【Still-Image MAD】What I Thought Was Unrequited Love (01:49; figures shown on the page: 5,870 views, 4 danmaku)
Buddies, put “NB”—fucking awesome!—on screen and in the comments!!
Clap-clap-clap-clap-clap-clap-clap!!!
And with that, the main part of this closing announcement is over, and so is Concrete Arcade: Still-Image MAD Festival 2019. Thank you once again to ZERO MAD TEAM, all the judges and all the creators. Your participation made this event a success!
Lab on Roof will contact every creator who participated in the festival as soon as possible to give each a small keepsake;
Lab on Roof will contact the creators of the award-winning works above as soon as possible to present the prizes offered by the event.
5. About Concrete Arcade, Lab on Roof and ZERO MAD TEAM

BY Yiban Youqu (一般有趣).
Concrete Arcade is the slogan for this event; the Still-Image MAD Festival is what the event consists of.
Concrete: the material, and the sense of giving tangible, specific form.
Concrete, a symbol of the metropolis, lines standing still,
Concrete and arcades condense the contradictions of modern society and give rise to superhuman fantasies.
Lab on Roof: ACGN arts criticism—animation, comics, games and novels—with an emphasis on animation, comics and games. A society on the roof for the study of contemporary audiovisual arts. We hope to bring readers more residual categories and insights while remaining humble.
ZERO MAD TEAM: Zero mad team is a MAD group focused on developing story- and emotion-driven works and “misinterpretation” MADs—edits that reshape existing material to suggest a different story. It emphasizes the importance of insights drawn from life and experience of the arts to creative work.
We know this event still had plenty of shortcomings. We ask the creators and viewers to bear with us, and we’ll try to do better next time.
Well, words have to end somewhere. It’s time to say goodbye—but we look forward to seeing you again!!!