Replacing pixel art

[Facet]

Ahhhhh

I had posted Alberto Mielgo's stuff here (Tron Uprising, Beatles Rock Band intro) and the servers went down nixing a few posts. I think this is the video Conceit mentioned I can't remember what my point was, but non-vector flat and painterly is pretty sweet, somewhat pixel like. I think I PM'd the links but forgot the thread.

Don't look up Mielgo if you don't want to see lots of boobs. 

[AlcopopStar]

I was gonna bring up the guilty gear stuff, it's just so absurdly cool. That kinda thing is the bigger "threat", if that's the right word, I feel it can get that crisp aesthetically focused vibe while having a lot more utility.

I see pixel art is something really close to low poly, something at the base of an art form that needs sharp attention to detail in order to work, I think it's best when it's utilizing it's strengths, low colour palettes, crisp edges, strong clusters, and worst when it's trying to be highly rendered digital art. But that's largely taste. Either way I think that toon shady stuff is a cool progression, a process more interested in general aesthetics and style then pure rendered detail, so there is some similarities cross medium there.

I guess whats new is that 3D a few years back realized it didn't have to look detailed or real in order to look good and now we are in this enormous blossoming of style. I'm not sure what that means for pixel artists, but I wouldn't be surprised if someone developers a shade or technique to mimic the look.

Reminds me of stumbling into this IRKALLA fanart from Mr Howard Day.


official

fanart

I think it takes most aesthetics of the qualities of the original and transfers them pretty well without losing much, of course the pixel art would be faster, right until you got to the animation at least.

That's the thing really, 3D always always has the upperhand in animation (and dynamic lighting I suppose). They haven't quite gotten traditional animation beat yet but even then the best traditional animation is often 3D assisted in the complex parts.

Anyway, food for thought.

edit: one thing i'll add about the guilty gear stuff is that the mouth animations look absolutely atrocious. one thing they haven't mastered I guess.

[Ellian]

I'm gonna drop this here: Ghost Trick - Phantom Detective

Pretty sure it's 3D animation (and it's crazy good), but the game really "feels" like pixel art.
I'm not sure how they achieved that, I guess they just really had a gread render... *shrug*

And on the NDS screen it just looks fabulous.



(Also the game is amazing and everyone should play it)

[tim]

It's actually very simple to 3D render to pixel art. Almost no shading / lighting, readable shapes, no shape smaller than 1 px on the scene, and it just simply works.

[Conzeit]

whoa! I kinda gave up on this topic picking up steam again, glad to be wrong   

@Facet heh, I feel like that's a "classic" now. Definitively still awesome.
@Elian @AlcopopStar I'm glad you guys brought up  Ghost Trick - Phantom Detective and howard's IRKALLA stuff, I remember seeing em and liking em.

@Tim: Glad to hear that! heh, that sounds like what's going on in Ghost Trick, but howard's IRKALLA fan art and all the GGX tricks ( I'm talking about HOW they achieved the aesthetic, which is described the behind the scene links), Paperman, hell, fucking Journey if we're gonna round up examples...they all give me a lot of hope that basically any aethetic can be done well in 3D, although, clearly it wouldnt be as easy as a base pixelart aesthetic.

@Alcopop, I like the discussion you bring up, about pixelart being like lowpoly and I largely agree, it's much of the reason I finally started to branch out a bit, when you see stuff like these games and David Oreilly shorts

 you realize that it's about picking an aesthetic with strict rules, and in there you can find incredible variety and many new forms of expression for each set of rules. That provides that feel of making pixelart that we love, and it can be done anywhere and with any selection of tools. Actually,here's a really interesting essay Oreilly made about this, I feel this is what pixelart is about beyond pixels

Where I see your point is when you say that really detailed pixelart is less interesting.
I see your point because I feel that when it's too much about detail and it starts to not show the tells of the technique it does become a bit like a really hard to do version of what people are already doing in ultra realistic photoshop images and in AAA 3D, that on top of that doesnt have a chance of being as detailed as either of those two things, because so much time is being spent on overcoming limitations instead of just adding detail.

 The demoscene pixel pieces that were done with "dirty" tools and were almost exact copies of photos in magazines would be a good example.

But I dont exactly agree because I think many different aesthetics can be done in pixelart, and it's possible to think it's just flat colors and squary looking stuff, like discussed in the Modern day pixelart? and  Uncanny Valley and Cuteness topics.


Everything starts getting homogenized to the same Sword & Sworcery aesthetic ( Darkfaxlz calls it hipster pixel XD ).

(Not that I think there's anything wrong with that look, it's fucking awesome to animate and it portrays things very economically, but nobody wants there to be just one flavor of icecream :p)

I think what's key is what Oreilly says, being coherent to whatever rules you started with, when you do that your piece will look good regardless of medium. (for example in the uncanny topic, I think much of the problem is we have engineers putting realistic faces with less realistic movement or behavior, or unrealistically cute faces with realistic details, the incoherence is what's uncanny)

What is key to me, about what I like in pixelart is that whatever you portray is portrayed in the most economic way, that direct focus on what the artist saw as beautiful in whatever is being rendered makes me feel like I'm being shown beauty, instead of just replicating in a phenoenom in a more mechanical way.

The more complex the phenomenom is, the more elements you'll have to bring in, but if there is still less elements than in the real thing, I still get that feeling of being shown beauty by an artist instead of having the object replicated.

 The water in this Fool piece has that, but his art in general is a bit too detailed for my taste, kind of like most of Metal Slug pixelart. They're both incredible anyway :p I guess it must be zoomed in for me to like it XD

I guess another way to put it is
I want you to show me what you felt when you went there, I dont want you to bring me there.

[AlcopopStar]

I largely agree with you, especially about the aesthetics being a set of rules, and David OReilly is a great artist to bring up.

I guess what I'd say to that is that while pixel art is applicable to all sorts of different styles, or sets of rules if you will, it has, as a medium, specific strengths and weaknesses that make some styles and rules more successful then others. Now i'm not trying to lay down hard and fast truths here, but generally I think we can look at PA from four different approaches in order to examine these strengths.

Artistically: Now, this one is a little redundant, but i'll say it anyway; anything goes. I want to make it clear that I am not trying to lay down the law on what pixel art is and isn't, I just want to make some observations, and if anyone is just creating for their own self satisfaction, then an objective framework isn't going to mean much, that aside we can begin to look at PA in more utilitarian ways, through both aesthetic success and economical (time or money) feasibility.

Historically: First however it’s always good to touch on historical context. If aesthetics are a set of rules, then we can see early pixel art as inherently having there own, perhaps unintentional, definitive aesthetics (due to the technological limitations). C64 restrictions for example, while it's an odd set of rules in the abstract it undeniable has it's own specific look made widespread through the necessity that the technology of the time demanded. It’s important to note here however, a lot of early pixel art wasn't trying to look like pixel art, a lot of the time It was trying to look as high resolutions and high fidelity as possible. This isn't true across the board of course, many games and companies found their feet with the art form aesthetically quite early, but I feel it’s still important to note, as many techniques that some feel are integral to the medium, AA and dithering for example, have historically been used more as an attempt to hide pixels then to express them.

Aesthetically: I view pixel art as a minimalist purification of digital bitmap based art, it stripes digital art to it’s literal base component pieces, and gives you manual control over where they go and how they are treated. In this way it gives us an opportunity to examine digital art in a stripped back fashion, what happens when we remove a million colours and leave 8? what happens when we take full control over the specific shapes and blocks of colour in any given place (clustering). The answer to me is we can create really aesthetically powerful things. It requires these aesthetic rules as a way to differentiate it from general digital illustration, and it has these rules encoded in it’s history. In this way it lends itself to an aesthetic treatment in a the same fashion that low poly does. Low poly is strong because it’s lost it’s capacity and hopefully intention to successfully be high fidelity or “realistic” art. Instead it simply tries to look good with what it has. This is why modern low poly often looks so great in comparison to your typical muddy modern fps. It focuses on colours, it pays attention to dynamic lighting, it relies on making the most of its shapes and textures. It needs to do these things in order to be successful, as it can't rely on looking realistic as a means for graphical success. There is a similar history here with pixel art, there is a reason early low poly looks, in general, so awful. It’s because a great deal of it was trying to look more rendered and more realistic then it was actually capable of being. It was stuck in that graphical rat race which valued looking real over looking good. Fortunately that emphasis has flipped these days and in that way I think we have seen a rise in styles that emphasize looking good, over looking real. That “hipster” pixel art you mentioned is one of them.

This is not to say that rendered pixel art is necessarily bad. There are plenty of pieces that look wonderful, and I think to a large extent it’s the specific treatment that is important. What I am saying is that pixel art is at it’s most powerful and relevant in a given image when it is not visually interchangeable with another medium, in particular digital illustration. Works where the pixel is integral to the art are more qualified to be pixel art over something that uses pixel art as a means to achieve a higher resolution look. Perhaps a good comparison is the way same way a pure photo-realist oil painting might be impressive, but the medium is ultimately more associated with a painterly, even impressionistic look; A look where the brush strokes and paint play a large role in the aesthetic of the image itself. Pixels are our paint. Therefor I value work that expresses them more highly then work that attempts to hide them. Works with large colour ramps or excessive dithering do note lend themselves well to the medium. This does not disqualify a rendered approach but does place it within the bounds of a set of aesthetic rules that attempt to utilize the pixel as the primary means of expression.

Economically: The last, and perhaps most practical viewpoint is one of economics. What can pixels achieve that in a given time that another medium cannot? Obviously, one of those answers is a pixeled aesthetic, however, and I think that, and this is the crux of this thread, pixel art can lose that territory, mainly to 3D art, which through forms of rendering, can take the pixeled aesthetic while retaining the incredible power that 3D offers (in particular dynamic lighting and animation). Ghost Trick is a good example of this, and while it hasn't mastered the approach it’s getting pretty close. Now not everyone has a Ghost Trick budget to work with, and 3D is expensive, but I think it’s our job as pixel artists in a new era to discover areas of aesthetic expertise that other art forms cannot easily claim. to some degree I think this is why the “Hipster” aesthetic of Hyperlight Drifters has been so successful; it is minimalist, colour focused, obviously pixeled, blocky and simple to create smooth animations in. To put it simply, it’s good territory for pixel art.

So with all that in mind I feel that as I said before, at least to me,  pixel arts primary qualities are low colour palettes, crisp edges, strong clusters. Elements that accentuate and complement the pixel, which is after all, the prime element we are working with.

What all this means to any pixel artists working out there, keep up your draftsmanship, it crosses into all mediums. And if work gets a little tight, maybe look into low poly, it combines well with pixeled textures and shares a lot of aesthetic similarities.

[RAV]

We've been comparing here the traitmarks of artworks. Another thing to compare is the very work of art. What are you doing? And when we look at the identity of the deed, are the kind of difficulties not part of that identity? In an absolute sense there is no replacement for pixel art; it is what it is.

A certain way of crafting has advantages and disadvantages, and this is what guides the art, different chances for certain decisions that go a different kind of creativity/solutions that lead to a different outcome. If doing something feels unnatural/inconvenient, you look for another way doing it or prefer doing something else more in its place.

Thus the most iconic styles of pixel art, in popularity representing the medium, are those that are most effectively expressed and efficiently made in that medium, according to its advantages and disadvantages -- this, on some level, any audiences understands and takes pleasure in; it looks interesting when there is a clever application of style to situation; smart is beautiful. Clever solution to constrain can keep up with unproblematic perfection; imaginative application can keep up with foundational originality.

So the question is, when you look to "replace" pixelart, for example with something 3d:
do you want to eliminate the problems pixel art is known for by creating it entirely differently?
or do you want to inject problems into pixelart that you solve in the patterns it is known for?

to be answered by and for yourself. Joy and beauty be with you. Power to the pixel.