What Are the Merits of Pixel Art?
"Pixel art" to me has always been a cute way of paying tribute to old games (or new games that just aren't using the latest technology yet). However there seem to be quite a number of people who take pixel art very seriously. To them it's more than just images with low resolutions and limited color pallets, it's actually an entire art form. This seems funny to me since basically everyone involved in making and playing video games has always been trying their best to dump sprite based graphics. Even before that was even feasible, they gave their instruction manuals references and their promotional posters big hand drawn pictures so players knew what each sprite was supposed to be, or at least help them believe that the tiny pixels on the screen were actually a huge monster. If it was true (or acknowledged as true) that the sprites were sufficient by themselves then that wouldn't be necessary.
This got me curious about how seriously pixel images are actually considered by artists. My first instinct is to check Wikipedia, where I found this uncited definition:
Purists within the pixel art scene hold that "true" pixel art should only be created from tools that place individual pixels (such as the 'pencil' tool), and that pixel artists should avoid all other tools including line, Bezier curve, circle and rectangle. Others counter that tools such as line and bucket-fill are acceptable as their functions could be just as easily, if not as quickly, replicated on an individual pixel basis.
Because of this rule, image filters (such as blurring or alpha-blending) or tools with automatic anti-aliasing are considered not valid tools for pixel art, as such tools calculate new pixel values automatically, contrasting with the precise manual arrangement of pixels associated with pixel art.
So what qualifies as "pixel art" is important to at least one Wikipedia editor. And the only real qualification he gives is that it needs to be all done by hand. This seems like a pretty superficial definition to me. (I'm sure there have been plenty of hilariously heated discussions somewhere on the internet regarding what tools are and aren't allowed.)
So are there any serious artists who acknowledge pixel art? I wouldn't be surprised if there are a few, but I admit I've never heard of them. Doing a quick Google search reveals nothing more than screenshot galleries and resources on designing sprites yourself. As far as I can tell, pixel art is nothing more than a niche drawing style by people who love video games, and a necessary technique by people who design those games.
This brings me to two possible conclusions. If pixel art is a completely worthy form of art then the fact that it receives no professional or academic recognition is a strange problem. But with games becoming more mainstream now this may change, especially if graphic artists deliberately choose to continue using sprites in this age of photorealism. The other option though is that pixel art is exactly what I first thought it is, a cute tribute to games that have to use it and people who take it seriously need to chill out.
One famous example of deliberate pixel art is Cave Story, which interestingly enough is being redrawn in a higher resolution by its creator, Daisuke Amaya. The revision is still pixel art (even by Wikipedia's uncited definition), but it demonstrates how the artist and community acknowledge that the original work could be improved (at least now that the improvements are available). I would like to point out that if they doubled the resolution again and redrew the sprites at even higher detail then it would look even better. And if they did it again, better still. Eventually using sprites would be inadequate for huge resolutions and they would have to use some other graphics technique. Of course by then it would be completely different graphical style rather than an upgrade and would ruin the effect the game is going for. But my point is that there was no specific artistic reason (excluding technical reasons) why the original sprites were drawn in the color pallet and resolution they were, and it seems immediately obvious to everyone that a higher resolution is automatically better.
Okay, the argument in that previous paragraph is flawed since it can't be denied that there is a unique aesthetic effect that images have with a lower resolution and limited color pallet. Part of my point is that nothing notable has formally been written on this effect, at least that I know of. And I could very well be wrong.
« Awards That Matter (sort of) Two Unrelated Things That Happened This Week »
22 comments
I play an MMO of sorts called Furcadia( www.furcadia.com ). You can make art for it but the editor for the art and the fact that it's 16-bit make you do it in what could be considered pixel-art style.
I, and many others who play this game, consider pixel art as art in which you have full control over each pixel. Line and buckets are okay, but anything that assumes what the pixels should be on it's own is a no-no.
The appeal in the art is probably more for the artists than the viewers. It's an art form where you take limited tools and use them in exacting ways to produce something bigger than the details. You have control over every little dot on your image, every minute detail. It's interesting compared to other styles in that the outcome is precisely designed bit by bit.
It's style is nostalgic to many gamers, but also enjoyed by the artists themselves.
While I've never read anything formal on the matter, Pixel Art has been around for some time, now. It's still very new, but it's been around for long enough so that some people could find converging points in the definition where they all could agree on. Heated discussions sometimes occur, but it's rarer than you think, mainly because of the whole "pixel consciouness" thing you mentioned. There are loads of tools that automaticaly dither you work (the process of using patterns of two colours to give the illusion of a wider colour range), and that's usually tought to be ok, since it's only saving you time.
Most of the "rules" (which mostly aren't rules at all), like trying to keep your colour count low, comes from the environment where pixel art was built in: videogames and computers restrictions. It's not uncommon to find artists trying to bend the EGA pallete, or C64 colours, or even the hideous MSPaint Pallete into something pleasant, and they usually succeed. People usually try to make every pixel and pallete count, because if you were trying to get whatever you're drawing inside a cartdrige, you'd want it to take the least ammount possible of memory, so that you could make your gamer longer, or hold more graphic tiles, or whatever. In that way, pixel art, regardless of what your definition may be, is totally and completely different from the others styles of painting. Even when people go for scruffy pixel art, there's usually a big attention given to...it's not even attention to details, but attention to the basis of the image itself, which, by itself, is wondrous.
I can only hope you get even more interested on the stuff and write something formal on the subject. This march, I'll begin majoring in arts and uni and am willing to write my final essay on Pixel Art. Having any start point of any sorts would be great :)
Finally, I'd like to leave you a link: http://www.pixeljoint.com , a online community devoted exclusively to pixel art, with various kinds of artists there, from pros to newcomers, from people that dealt with the Demoscene community and all kind of stuff. Heated discussions about high colour count pieces guaranteed. Might be and interesting look from the inside, if you're still interested.
wow, there are some serious spelling, grammar and even logic mistakes in my comment above. I should stop trying to write right after waking up.
Okay, this is my last comment before an answer, I swear. I don't want to clutter your comments :P
http://pixeljoint.com/pixelart/39290.htm?sec=date this is something most people wouldn't think of pixel art at first glance. But then again, one question commonly asked in pixeljoint to artists that go crazy like that is...if it could be easily done in CG, if it doesn't really remember you of pixel art, why should one pixel such image? This brings back your point of Tribute vs Style. If it's got to remember you of the past, well, then it ain't really much more than a tribute. But that kind of tought isn't shared with everyone, and that's where confusion settles in.
Also, something I forgot to include in my first comment, the whole concept art thing is as true as it can be when looking at the NES age, but take a look at the sprites on Breath of Fire III, PSX. Those things pretty much stand alone, they're all pretty readable. And all below a 32 colour pallete. We don't even need to go that far, the SNES has had loads more of pixel art than the PSX, and people could mostly always make out what they were without a external visual guide. Megaman 7 comes to mind, but that's prolly because Megaman ALWAYS comes to my mind :P
You could have researched more about it, at least a timeline in comparison to the technology from that time, before "ranting" Pixel Art this way.
By the time Pixel Art was born, it was not even classifed as Art itself as its purpose was to solely make graphical interfaces using the current and limited technology form before. Not even digital art was fully recognized as Art too. There were no Pixel Artists, and that would only happen when Nintendo first released NES. Even then, how could Pixel Art compete with other digital art techniques that were evolving much fastly, such as Photoshop? A tool that were far more familar to the former artists than Pixel Art (which we can easily tell: it's hard to digest)?
Nothing helped an academic recognition for Pixel Art as it was only a mere need for the video-games market and technology, if not simple interfaces for home computers. The whole thing was only being used as an artistic expression more or less at the Amiga age.
And why not a professional recognition? See, I was hired not for being an artist, but for being a Pixel Artist, and I bet a infinity of other artists who works on celphone and Nintendo DS projects were also hired for this.
To finish it up, there is actually literature about Pixel Art, at least for the game developing area. A book that's called "Character Design for Mobile Device". The title does not fit its content though.
I admit I only did minimal research before writing this post and there are people who know much more about this than I do, but what inspired me to write it is the fact that good resources (at least that I thought were good) were difficult for me to find. (Something takes longer than 10 minutes to find something with Google and Wikipedia = too hard for me). Pixel art just seems to get very little recognition as a serious art form or a technique compared to other visual media, and just stays in a small niche.
I want to clarify that I didn't write this entry trying to make a point. I don't hate or love pixel art, I'm genuinely curious about how serious it is taken. This post was more of a stream of thought rather than a serious argument.
And I appreciate all your input so far, I'm learning new things!
I blame the world!
There are some artists who create pixel art unrelated to video games. I guess this is mostly in the graphic design / marketing field and not proper serious art, but that's still quite removed from the video game stuff.
I only remember eboy (see: http://hello.eboy.com/eboy/about/) right now, but there are more graphic design studios who have done this kind of work.
Here is a rant about the definition of Pixel art:
‘Pixel art’ is a self important term. It is used to describe art that could run on crappy – I mean, retro machines (eg. Nintendo, amiga, c64). It is also used to describe art that has a similar appearance to the aforementioned art. That is, small size, efficient use of colors, clean looking (eg. antialiasing to smooth lines, dithering to neatly transition between colors), detailed art (eg. highlights, texturing - at low resolutions areas of flat color looked bad imo). There is no correct definition for how similar they need to be. What one person believes is an efficient number of colors to classify a piece as pixel art another person may completely disagree with. Someone who grew up with a c64 (16 colors) might find a 256 color image excessive (though this was possible on amiga AGA).
The current status quo at pixel joint I believe is to use a methodological definition – “place each pixel by hand”. This definition is not only superficial but is flawed. Firstly, unless you sit behind someone as he draws, you don’t know what you get – hence it is an unworkable way to classify art.
The definition is incomplete since a secondary definition is used at Pixel joint. Since messy, unappealing art can be created which conforms to the method definition the rule to exclude art is “it’s too WIP”, which sometimes is really saying “this doesn’t conform to the appearance of retro art”. However, it is a methodological definition first and foremost. If you produce art that fits the retro look but does not conform to the methodological method (e.g used color reducing or brushes to achieve that result), then it is excluded. I’d like the method definition to be removed at pixeljoint and if any academics write bullsh*t about pixel art hopefully they don’t use that definition either. It is harder for the lay person to understand since you don’t get it unless you actually draw a piece placing each pixel by hand.
In some ways though, the method definition makes sense – a high level of control facilitates neatness, detail, efficiency. Ultimately though, it hinders making good realistic pixel art (which was a goal of some games on retro machines).
High level of control isn’t always a boon. If it were, people would always use a mouse or keyboard to pixel, rather than using a wacom tablet. Using a wacom, you can create smooth, beautiful lines, which is not as easy with a mouse. The current definition, by insisting on a high level of control AT ALL STAGES, does not facilitate good form (ie. good lines or shapes) and does not allow artists to take shortcuts to produce art that still fits the pixel art ‘aesthetic’. Its like the people who advocate the method definition don’t want to draw good pictures and don’t think our time is valuable.
The justification for the methodological definition is flawed. The justification seems to be that by ‘consciously placing each pixel’, we are making better art. However, I don’t believe art is a process where you need to make conscious decisions (at least not all the time). The part of your brain that likes to come up with definitions and make decisions and crap on like I am now is separate from the part which allows you to draw. Just like driving the car you might ‘zone out’ (ie. not remember driving from point A to B because you were thinking about something important but somehow you obeyed the speed limits and didn’t run people over) you can ‘zone out’ drawing. It’s a repetitive process – a monkey could be trained to do it.
The second reason this justification is flawed is because if conscious decisions do take place they take place at the planning stage but placing each pixel by hand does not help these decisions. It is harder to know what elements to include and to get the composition right because placing each pixel by hand is time consuming. Hence the method definition does not result in better art nor art that is made in a more conscious way.
In some ways a methodological definition forces an aesthetic. If the rule was everyone must use a mouse, possibly more art would feature a “jaggidy” line style rather than smooth lines (eg. the very squarish guy in the gods intro pic). But is it right to insist that everyone uses a mouse and exclude art that uses a wacom? It’s effectively saying “jaggidy” line style is correct but smooth lines are not. Why would “jaggidy” line style be more correct? This is why I see the current definition as a result of people who like Eboy type art (simple, stylised) not acknowledging the merits of realistic pixel art, which was a goal of many games on the retro machines (see ‘Defender of the Crown’).
I can see three alternative definitions for pixel art:
Pixel art is art that
1. could be displayed on retro machines (eg. c64, Nintendo, amiga etc).
2. is created using a tedious method of placing each pixel by hand
3. is clean, uses colors efficiently, detailed, may resemble art that can be displayed on retro machines, <256 colors, <800*600 pixels size.
Three is would be best for me at least. I don’t think it is adopted at Pixel joint because communities are a bunch of people and people in bunches are stupid. Hence the definitions made by academics would be good (provided they agree with me ☺ ).
Art is never well defined though by a simple definition:
"The intellectual is constantly betrayed by his vanity. Godlike he blandly assumes that he can express everything in words; whereas the things one loves, lives, and dies for are not, in the last analysis completely expressible in words. "
Author: Anne Morrow Lindbergh
Would you have found 3. more useful in the Wikipedia article John?
In terms of merits of pixel art it could be time considerations. I think it’s quicker to draw a animated, detailed low resolution sprite than high resolution sprite. More detail means more things moving means more work.
Notalgia and restrictions (eg. ds, mobile) would come into play. Lastly, pixel art can be easier for a poor artist to pull off rather than a higher resolution sprite.
Pixel-art rulz for many reasons... it's art, is cute, is useful, is retro, is colorful, is SNES, is authentic, is nostalgic, is free, is MegaMan, etc. But I just wanna say that I love it cause is a non-common art that could be used as a hobby and surely (if well done) could astonish to everyone. Sometimes is relaxing to place those pixels and fell you did it in the best way.
Pixel art means the effort to make the things right no matter the time they requires. It's a kind of perfeccionism art. Besides, pixel-art is frequently usend in many ways, for example: videogames (off course), emoticons, websites, publicity, etc ...blah blah blah
And excuse me if "All your base are belong to us!"
I never got this whole argument of mouse vs tablet. They are just tools. I know people who pixel on the C64 with a joystick or the keyboard, and they create just as smooth lines and AA as I do with my wacom tablet. It is just what you are comfy working with.
As far as I am concerned pixelart is digital art where care is taken on the pixel level and there is control over every colour as well as the placement of each pixel. This does not mean that we need to place every pixel one by one, but it means that the machine does not make decisions for us, as it happens when you make a brushstoke in Photoshop. Lots of colours and antialias gets generated automatically.
As long as you get to something where you had full control over everything, that is pixelart, tools like fill, the linetool or whatever are fine as long as they do not create any additional artefacts which are not perfectly foreseeable. Then again, that's just my 2 cents.
I'll put it like a friend of mine likes to put it. Pixel Art is digital mosaic. That's essentially what it is. Laying down tiles of color meticulously to make a larger image.
I don't think it would be a stretch to say MOST pixel artists got into the art form because of games. Some pixel artists go on to create professional pixel art for game studios and advertisements. The form has evolved. It has its own merits and its own techniques and its own style. Making pixel art is about knowing how to exercise CONTROL in your artwork, far more than most forms of art. The difference between great pixel art and poor pixel art often comes down to how well the artist was able to conserve colors and still represent the intended image, and it doesn't take a trained eye to tell the difference between high quality work and low quality.
Unfortunately a lot of people who do pixel art don't really take it seriously, or try to improve. Maybe what a lot of the world is seeing when they look up Pixel Art is the kind of crap that surfaces in the DeviantArt pixel gallery. It's an art form without a whole lot of coverage, and the community of serious artists who actually strive to get better and produce professional quality work (and beyond, I've seen hobbyists do much better artwork than some professionals) is confined to a small corner of the internet.
There's no reason to dismiss pixel art because it's an underdeveloped form ar artwork though. Like I said, it's a unique style that requires an artist to hone an entirely different set of skills than painting or drawing, be it digital or traditional.
I agree with ptoing. Pixel art in it's simpliest definition in my mind is
Digital art created in a way that the machine does not make any decisions for you.
Ptoing - but the question is can you take that photoshop line, color reduce it, remove artifacts as you see fit, and then does that constitute pixel art? Is pixel art a result or a method?
I don't get this 'decisions' concept. When you draw a line using a wacom tablet, isn't your hand making decisions for you?
Well if i mean line tool of course I mean one without antialias. And yes, I think if I would make a picture in photoshop, scale it down, colourreduce, tweak the colours and tweak every little thing so that in the end I had full control over the thing, then yes, it would still be pixelart.
So you really believe there is a difference drawing with a tablet or with a mouse? My hand is leading both, my eyes see pixels on the screen. I draw a line, or a blob of colour, I look at it, I like it, good. I don't like it, I work it more. My hand is not making decisions, it is not a thinking entity, I control my hand. I made pixelart with a mouse, and I still can if I want to, and rest assured the outcome would be the same, the only difference is that I am about 20 times faster with a tablet and working with a mouse is like moving a brick around on the table. A stylus is closer to a pencil, it is a more natural extension of my hand than a mouse when it comes to drawing.
By "the machine makes decisions" I mean, that when you work with tools like blur, transparent brushes and the likes in photoshop you do not have a 100% foreseeable result. If I work in Promotion, DeluxePaint or the likes and set a ramp to then work with shade mode, where when I paint over any colour in that ramp it goes one step up or one step down, that still is 100% in my control. If I take a ditherbrush to place some 50% dither somewhere, that still is 100% in my control. Why should I feel obliged to put down every pixel of that dither pattern down myself, when I know that it will all be xoxoxoxoxo. I could do that, but it would take longer and I have other things to do with my time than just place large fields of dither by hand. I did that, I know how it works, I know how it will look like if I use the ditherbrush, so that means that I have full control.
Imo pixelart is both a result and a method, it is an medium in itself in my understanding and not just a style either. If you take some other medium, say acrylics, someone could go and put it on the canvas straight ahead, painting free. He could use lots of stuff like masking fluid, stencils and whatnot. The result would still be an acrylic painting.
Now with pixelart it is a bit more complex as everything on a computer screen is pixels. As I said before, as long as the artist had full control over what he did and there is full attention paid to the pixel level, it really does not matter what went down inbetween. If you take a drawing you made and scan it, resize it, and then go over that in any package with the tools for pixerart, bam, that's pixelart, don't you agree?
So really, what is the difference if you paint the first stage in photoshop and then shrink, colourreduce it and whatnot. As long as you go over it, revise the stuff the automation did to it you are still fine in my book. A good artist would know how stuff scales down and how much rework it would need to get it up to polish.
So in short:
Digital images which were worked with attention to the pixellevel on a holistic level = pixelart.
Ones that were only partly worked with pixellevel detail are not pixelart, but hold elements of pixelart (this you can see in many 3D texturemaps)
Then there is index painting, which is kinda like photoshopping at low resolutions, but then often also polished up with a layer of pixellevel detail. This then could come to be full pixelart as long as no parts are left where automated aa or whatever is going on.
At the end of the day, this is something I feel many people will always disagree about. Some people out there claim whatever is made with MSPaint is pixelart. Then you might as well say most oekaki is pixelart, which is not the case. While oekaki can be pixelart, not all oekaki is, and pixelart is not oekaki. But, I am rambling, I should go to bed.
I like your explanation alot. This part especially:
"Well if i mean line tool of course I mean one without antialias. And yes, I think if I would make a picture in photoshop, scale it down, colourreduce, tweak the colours and tweak every little thing so that in the end I had full control over the thing, then yes, it would still be pixelart."
That's how I did these two. No-one seemed to notice (or maybe if they did notice they didn't care).
http://www.pixeljoint.com/pixelart/38491.htm#
http://www.pixeljoint.com/pixelart/39335.htm
I think using a wacom is fine, I only brought it up to further the point that placing pixels by hand definition is flawed.
To avoid discussions I think pixel art is what you can do with MSPaint or 1x1 Lego blocks.
I'd just like to point out that something very close to pixel art has existed for millennia.
Mosaics!
i personally believe that pixel art shouldnt be defined by the canvas size, what tools you use, if it was placed square by square or fill tooled in. what i believe makes pixel art... pixel art is your color palette whether it be basic 8bit (black, white, red, blue, yellow, purple, orange and green.) 16bit, 32bit, and if you want to push it 64bit. further more each pixel should be a pure color, meaning there only exists a single color in a single pixel. theoretically everything is pixel art on any machine they all form pictures with pixels, tvs, computers, cellphones, etc. they all have pixels. hence why i believe they way i do if you can use an enormous color palette then its not really pixel art but using a limited amount of colors to make the same picture that uses a bazillion colors, now that truly is pixel art