Recently, Dinofarm Games' Blake Reynolds wrote a blog post in which he and his studio renounced pixel art. I'm here, essentially, to not only further the discussion of what pixel art can be within the medium of video games, but also to accuse Reynolds and co. of picking up their toys and going home after someone was a little bit mean to them.
The article does a very good job of explaining why pixel art can be beautiful and unique, and I highly suggest that you read it if you haven't. Take a look especially at his Street Fighter 3 versus Street Fighter 4 example, but suffice it to say, you can create some amazing animations when working in 2D that are much more labor intensive to recreate in 3D.
Now, the thing is, you could also do 2D art with illustration or vector art, and it would be much more scalable. See Skullgirls for an example of excellent illustrative 2D art. New tools are also popping up around the industry with the aim toward making simple 2D animation easier. See UbiArt in Valiant Hearts, or Spriter in our own game Gunhouse. So if this stuff is easier to scale, and easier to make, why does anyone work with pixels?
Well, here's an analogy for you: Why does anyone still use oil paint? Oil paint is unwieldy, it's hard to master, and you must commit to your decisions. You could get a much higher level of detail with greater ease using photoshop (that undo option!). But people still learn oil painting as a craft. It evokes a different style – it gives the artist a feeling of being close to he work. The labor involved is part of the craft, and is visible in the result.
Now, I agree with Reynolds that in the past, when 3D was not an option, game artists were just trying to do the most HD thing they could. But as soon as 3D technology became readily available, let's say during the PlayStation era, it immediately became a choice. And not only that, for much of the 90s, pixel art was still capable of more detail than most 3D art. If you look at the 2D representations of characters in Japanese games of the era, especially, they're more symbolic rather than realistic. Something Pixar does today to excellent effect.
While even the 3D of the era was much more scalable than pixel art, representations of humans and organic forms were much more detailed and relatable in pixels throughout that period. (Ignoring the technicality that all game art is ultimately displayed as pixels).
Street Fighter EX Plus Alpha, versus Street Fighter 3, both released in 1997. Which looks more like it has humans in it?
So, why did pixel art continue to exist, when more high definition technology existed? It's a choice, made by artists who are drawn to a certain aesthetic, a certain workflow, a certain connection with what they do. Pixel art is a technique – it's a tool. So is 3D. You can choose to use one, or both. Even in modern 3D animated features, many animations are first done by hand, on paper. Certainly, in the 90s, some artists stuck with pixels because it's what they knew, or they became texture artists, which uses many of the same mental muscles as pixel tiling.
But now, it's a choice of art style or technique. It's a choice that's as legitimate as choosing to paint with oils over acrylics, or watercolor, or photoshop. People who create pixel art, real pixel art, do so because they love the artistic medium. And as a medium, it has its advantages and disadvantages. Yes, as Reynolds says, pixel art does not scale well without getting muddy, stretched, or otherwise molested. But on the animation side, try doing something like Paul Robertson does in 3D, see how long it takes you, and how far you get.
Paul Robertson's Kings of Power 4 billion%.
People like Paul do pixel art because they enjoy it as a technique, and as an art form. For us, with our game Gunsport, pixels are a very conscious choice. The game is visually and thematically of an era, namely the days of the 90s arcade. Does this sound like pure nostalgia to you? That's a debatable point, and my art director might accuse me of it – but consider films that use the visual styles and techniques of a bygone era for specific effect. A film like The Artist, in general, uses the production and filming techniques of the silent film era to evoke a space in time. Why shouldn't the same happen in games? If we choose to evoke a certain feeling, and make some of our production techniques match that era, is that not a legitimate choice that an artist can make?
There has been a backlash against pixel art lately, and this is what Reynolds finds upsetting in his original blog post. He finds that people don't appreciate the craft and intense labor that goes into pixel art, using “pixelated” as a diminutive statement. And to that I say, well, shouldn't we be teaching people about the craft, rather than throwing up our hands and saying “oh well, I guess the kids hate pixels?”
There are people who think Jackson Pollock is terrible. Frankly, I'm not a fan. There are people who can't stand Picasso. People certainly didn't like him that much at first. You could dismiss Mondrian as just some lines and colors.
I'm not trying to compare game art to Picasso, but I am saying that art is subjective, and what people like will evolve over time. And maybe you "don't like pixel art." Many have said this recently – but IFnd it a foolish statement. Maybe you don't like 2D games in general, and that's fair enough, but to dislike a technique, well, how can you?
Not all pixel art is the same, and not all of it retro. Superbrothers is not Cellusious, is not Paul Robertson. All are pixel artists, all have found their own style, and none of their styles existed before, in a "retro" sense. They operate outside of the constraints that made pixels a necessity in the past. Pixels alone do not mean retro. Again, they're a tool that an artist can use to create something interesting. That's it.
Superbrothers (top), Cellusious (middle), and Paul Robertson (bottom) all have unique styles that one cannot accurately call "retro."
What we as appreciators of pixel art should do is not just explain, but show how the technique is different and evolving. Reynolds' post does a good job of showing how it's different, and what's good about it, but at the end, while coming to the same conclusion – that hand-crafted, meticulously-created artwork is unbeatable – he gives up on his chosen form instead of forging ahead, and showing people why it's worthwhile.
Now, if Reynolds just secretly wants to try his hand at 3D, that's another story. But if you just want to go where the market is, and if you're in video games primarily to make money, you'd be better off making an advertising or monetization platform. If you love pixel art, do pixel art. If you want to do 3D, do 3D. If you want to get somewhere in between, as Arc System Works is doing, go for it. But stick to your guns about your art, and don't let a few consumer misunderstandings dictate the course of your creative life.
Guilty Gear Xrd uses cel shaded 3D to emulate the look of pure pixel art.
It's incredibly difficult to find good pixel artists these days. It's rather a lost art, and most of the people you'll find available for work either have a sloppy style, don't have traditional art training, or make art that only works on a static tumblr page, and not in the context of an actual game. But still we do it. We forge on with pixel art because we are trying to say something with it. It's thematically embedded in our game, even as we use a few new techniques to make it work. Is it limiting? In many ways yes, but straining against limitations, even self-imposed ones, makes for good art.
I'm not saying Necrosoft Games will never do 3D. We probably will! But we will not renounce pixel art. Does a painter renounce the brush, or the sculptor their clay, simply because photoshop and Zbrush exist? Or because people "don't get it?" Rather than renouncing pixel art, we should all work harder to prove to the naysayers that we've got something of value. That's what we will try to do on our side.