The Influence of Art Direction
Guest editor to Polycount, Nick Halme, brings us some commentary on the influence art direction has on design, and how design influences art direction. The goal with such an article as this, is to hopefully bring on discussion over the matter.
The Influence of Art Direction
When you get down to it, a game is about providing a service for players. That service is something more substantial than "fun"; it's fantasy. Every game allows you to step into that magic circle -- follow the rules of a new world, forget about your own for a while. But every time a game misses the mark and doesn't communicate clearly, you're forced out of that circle of immersion. Every time you're not sure how you're supposed to fight a boss, every time you can't tell where you're supposed to leap next, every time you're not sure how you died -- the game has failed. Maybe you even put the controller down. It's a problem that can be caused by any facet of game development, but here I'd like to talk about how art direction and design can best jive.
Player fantasy is one thing, but part of the player believing he's in a world is user experience. I don't mean focus testing or popping up help text every couple minutes (that's worse), but making sure players have all the information they need to win, or to remain engaged. A huge part of that is art direction.
Doom didn't really have art direction like we think of it now. The games of that bygone era were driven by technical restraints. In Andrew Hutchison's essay on Game Studies, Making the Water Move: Techno-Historic Limits in the Game Aesthetics of Myst and Doom, he describes how the two games had made vital tradeoffs. Myst had beautifully realized environments -- but it also operated, play-wise, as a point-and-click adventure. On the other hand, Doom sacrificed graphical fidelity for first person shooting. When these games were made, people just didn't have the space to think about things like communicating game mechanics through art style -- they were focused on making the damned things fit on however many floppy disks and run on your 66MHz processor.
[caption id="attachment_3399" align="aligncenter" width="644" caption="id Software's DOOM"]
[/caption]
So, with that in mind, we can say that the Revenant didn't look the way it did for any particular reason. How does a seven foot tall robo-skeleton communicate to the player that it's about to shoot rockets at them? While I'm not against seven foot tall cybernetic corpses, it looked the way it did because that was pretty fucking cool. There's nothing intrinsically wrong with that sort of natural, original art -- but it can coexist with purposeful art design that tells players something about its function. It is art for a game, after all.
A good example of an art style that still ignores the communication of gameplay is one driven by the most popular player fantasy: Military shooters. Games like Battlefield and Counter-Strike have one priority, and that's to make players feel like they're a hardass commando with an itchy trigger finger. The problem is, it's maybe not the best experience to be shot in the face for no reason -- target recognition is something these sort of games really struggle with. Is the guy with the red hat on my team? Why is he my enemy if we're both wearing snow camo?
Call of Duty dealt with communicating to players what's happening very overtly: They implemented the "killcam", which gives you an instant replay of the degenerate hiding in a bush who put that bullet in the back of your head. This works to a degree, but the player is getting information after he's already been killed -- and it's a mechanic that exists outside the game world.
[caption id="attachment_3422" align="aligncenter" width="650" caption="Infinity Ward's 'Call of Duty: Modern Warfare 2'"]
[/caption]
Games like Team Fortress 2 and the upcoming Brink are visually interesting for a reason -- they improve gameplay, besides making your eyes happy. TF2 was, in previous incarnations, just another shooter with army men. That's not to say the gameplay wasn't there, but it's clear that the visual overhaul added both heaps of character and mechanical clarity.
"Art direction should be closely tied to gameplay," says Art Director for Relic, Ian Cumming, in relation to Warhammer 40k: Space Marine, "but it goes beyond unit ID, especially in our game with its IP limitations. We’re still struggling a bit with unit role/ID because of this, and beyond silhouette and colour, have found that FX play a big role in distinguishing units."
Rather than chase reality TF2 created it's own world, where rocket jumping makes sense, the Red and Blue bases are absurdly contextualized, and the heavy weapon is always carried by a gigantic Russian man.
Quoting Valve's own documentation on their rendering techniques, "...the art direction and technology choices combine to support artistic goals and gameplay constraints. In addition to achieving a compelling style, the shading techniques are designed to quickly convey geometric information using rim highlights as well as variation in luminance and hue, so that game players are consistently able to visually 'read' the scene and identify other players in a variety of lighting conditions."
And besides, like Ron Perlmann keeps telling us -- war never changes. But if you want people to pay attention to your game, maybe you should consider changing what war looks like. "A game's visual style is your calling card. Without an appealing aesthetic, people may never be drawn in to experience gameplay," says Bryna Dabby, producer at Tiny Speck. But once they're in, compelling gameplay is what keeps people playing. Your game needs to look and play beautifully."
But more importantly than showcasing the need for design and art to work together toward the same goal, games with distinct and purposeful art direction are tugging games away from miming reality and toward making their own worlds and rules. Brink's exaggerated characters and colourful environments, much like Borderlands, lets us escape into worlds that not only allow developers to be more creative, but allow players to experience that new world as well. More big budget titles need to create their own source material. So it's an exciting time -- it took film nearly a century to create striking simulacra like Sin City or 300. In just over thirty years videogames have been given the technology to render incredibly realistic scenes that, like Heavy Rain, could very well mimic what we see in film. But what's exponentially more exciting are the unreal and fantastic worlds -- those worlds with the artistic license to integrate art with the fundamentals of a game more completely than ever before.
-Nick Halme, guest editor for Polycount
[caption id="attachment_3395" align="alignleft" width="62" caption="Nick Halme"]
[/caption]
Contact: nickhalme[at]gmail.com
Editor information: Nick has previously written for the Escapist, Gameswatch and other various sites. During the day, Nick Halme works at Relic Entertainment, by night he hunts the undead menace and by night he foolishly forgoes sleep in order to put words together for the internet. Follow Nick on Twitter.
If you would like to be a guest editor for Polycount, contact us us and let us know.
Replies
The Doom screenshot I see posted in-line does a great job of illustrating the bold colorfulness of the Doom world, far more strikingly composed on first glance than the MW2 screenshot which is what, meant to show our great and varied progress in our visual presentation for games? Or is it to set up that the Space Marine screenshot? The article bounces through a series of disconnected, debatable observations on the subject of art direction. What are you getting at? "Art direction is good and it helps." ?
I understand both you and Adam work at Relic and are justifiably proud about the games you are associated with, but it's hard to take this article as anything more than a little pimping out of Relic's upcoming game.
And what exactly are we seeing in that screenshot that represents what Relic brings to the table? All of the classic WH40k character designs--the bold colors and distinct silhouettes--are from lessons learned and implemented on tabletop miniatures, well before even Doom. Arguably we've forgotten more about great character design and compositional elements gleaned from classical tenets of illustration than the current glad-handing consensus about AAA art direction would suggest.
I also think it's pretty short-sighted to consider most any current action game as having character art built to much more rigorous standards than "looks cool," which is all you seem willing to ascribe to the Revenant.
Yes all this good to-do about silhouettes but have we really seen a monster set as easily, instantly identifiable as the Doom 1/2 bad guy set? Does the Revenant's design need to communicate anything more in the "about to shoot rockets at you" department than it already does, or does the fact that it will always shoot rockets at you within seconds of spotting the player suffice?
This is not a personal attack, I am just deeply disappointed for having just read that article.
I cannot stress enough that I am very impressed with the polycount redesign and the efforts to continue to write new content and get it posted, but come on guys. This is a first draft reaction paper for freshman comp level stuff, it's fluff written for I don't know who. Your credits posit you as a published writer--so who is your audience here? I find it totally inappropriate to the kind of shared expertise and craft level on polycount. We've got several guys actively working on Brink. For a community as deeply invested in game art production--if we're honest, one of the very best going, really--shouldn't we be pitching the level of discourse a fair bit more complex?
I agree Adam, although i'm no where as experienced as you in knowing what goes on in a white-board meeting, I can kinda see what you mean. If Art and Design communicated well during these meetings, they could essentially #1 fix a lot of bugs and #2 they could realistically see what they can and cannot do to prevent a ton of issues. I wouldn't mind watching some of these meetings somehow, but maybe someday ill be apart of them
This was an interesting read. However I feel the content is missing clear examples of what you're trying to communicate to the reader. Simply stating that game design and art direction need to work together doesn't really provide a solid platform from which stimulating conversations can be launched.
Reality is: anything you see in a video game is going to be there because someone thought it would be "really fucking cool"...
Nothing has really changed about that since DOOM. Designers come up with a crazy Idea for a boss fight experience and the thing tailspins from there most of the time
In my opinion the problem most games face isn't that the art direction isn't helping the gameplay, but that the narrative isn't thought out well enough or perhaps interesting enough to keep players immersed. You can play a dog ugly game and be really into it if the story is well written and the gameplay supports it.
It's only recently that art direction has been a huge focus in the developement of a product. This is because technologically you can do more than you could "back in the day" but mostly because it's actually become lucrative. Back when DOOM was made, you had guys and gals (mostly guys) who thought it would be cool to make a game (mostly programmers-designer hybrids) who had a buddy that could kinda draw stuff, working in a garage somewhere in BFE. There simply weren't a whole lot of artists (or even "artists") wanting to do digital art.
When developing realistic shooters, there doesn't seem to be any artistic way to communicate something that couldn't be communicated in reality. I mean, how else would you comminicate to a player "You got shot in the face by a sniper... you are now dead" other than a kill cam. This particular instance isn't one I think you can even begin to address with art direction.
This is the problem with trying to create a truly realistic experience. However, over the years we have trained players to expect certain things from a game, and most times they understand what happend without being told exactly "Sniper... bushes..."
This is no excuse for bad game design, or weak art direction, but something that we come to accept as game players... until we have something that more closely mimics reality in terms of interfacing with the device; we will continue to have these problems.
TF2 is good solid art direction.
However, it's next to impossible to solve (through art direction) the player driven problems existing with user experience in a multiplayer game; if the player is being powned by some exploitation of the layout of a particular map or poorly balanced game mechanic... these are solutions for game designers. Artists should play the game to help find these problems, but art direction doesn't fix flawed design (no design is perfect).
Art direction plays a much bigger role in single player games; something that people don't really talk enough about either. Nevermind that TF2 looks dope; and while it does a great job of communicating to the player what they need to know, there's no real narrative to communicate through visual stimulus there. The artists/art directors can throw in details here and there to let players come up with their own back story of the corner of a room, but the story has no impact on the gameplay experience. In a lot of cases, people turn graphic settings all the way down for competative edge. If they don't like how dark a game is because they can't see the characters they'll crank up the gamma, or turn textures off in really busy games, or even if they just have a slow machine... bye bye art.
If a single player game has shitty art, or art that doesn't make sense, or push the narrative, the player will lose interest faster than they would in a multiplayer game. But there has to be a strong story to tell before the art can even begin to have a chance succeeding.
Equally so with design.
I think the problem goes well beyond "why is he wearing cammo pants if i'm wearing cammo pants..."
One last thought dump; not knowing immediately where to jump in a game or how to solve a particular problem isn't a fail of the game if the motivation behind the design is clear. I think one thing that games have done over the years is training players to NOT think.
Looking forward to the day that it's no longer about telling the player where to go or what to do, but creating a world/gameplay experience in which players can explore the world in a seamingly natural way. The best experiences are when you're playing a game and think, "I wonder what will happen if I do this..." and it turns out to be something really badass and natural feeling.
Thanks for the article
The Doom screenshot I see posted in-line does a great job of illustrating the bold colorfulness of the Doom world, far more strikingly composed on first glance than the MW2 screenshot which is what, meant to show our great and varied progress in our visual presentation for games? Or is it to set up that the Space Marine screenshot? The article bounces through a series of disconnected, debatable observations on the subject of art direction. What are you getting at? "Art direction is good and it helps." ?
Arguably we've forgotten more about great character design and compositional elements gleaned from classical tenets of illustration than the current glad-handing consensus about AAA art direction would suggest.
I also think it's pretty short-sighted to consider most any current action game as having character art built to much more rigorous standards than "looks cool," which is all you seem willing to ascribe to the Revenant.
Yes all this good to-do about silhouettes but have we really seen a monster set as easily, instantly identifiable as the Doom 1/2 bad guy set? Does the Revenant's design need to communicate anything more in the "about to shoot rockets at you" department than it already does, or does the fact that it will always shoot rockets at you within seconds of spotting the player suffice?
@nfrrtycmplx - Your post alone made this article worth it. Great insight.
@gauss - Debatable observations is the point here
Assuming gauss is too humble to link to himself, I'll just link his articles on character design, which are very much related to this subject and also very nice, for him:
http://designreboot.blogspot.com/2009/11/visual-clarity-in-character-design-part.html
http://designreboot.blogspot.com/2009/11/visual-clarity-in-character-design-part_21.html
The Left 4 Dead blog had this interesting post: http://www.l4d.com/blog/post.php?id=2129
Heck, any time you can follow Valve's in-game commentary, do so, because that's all priceless information.
Additionally, the idea that pre-TF2 Team Fortress predecessors were "just another shooter with army men" is outrageously inaccurate. Team Fortress for Quake 1 may have lived on modified Quake 1 assets, but the game communicated it's stylized depiction of team based shooter classes so successfully that it spawned incarnations across almost every popularly modded FPS engine. The culmination of which being Team Fortress Classic (on Valve's own Half-Life 1 engine), which DID have highly distinguishable character models for its classes that were far more stylized than realistic.
Just because they weren't some advanced form of cel shading, or did not have oversized hands and jacked up proprotions does not make them less successful at visually communicating their role in the game. These characters communicated their roles just as successfully with far cruder visual capabilities (1,000 triangles for a character? If you're lucky).
Art direction meshing with design is NOT exclusive to wacky or otherwise unrealistic art styles, to suggest such would be a ridiculously ignorant view point, and to argue that it is in any way the best or only route to successful player guidance is insulting.
"Doom didn’t really have art direction like we think of it now. The games of that bygone era were driven by technical restraints." Is NOT the same as, "Doom didn't have art direction, it was all a fluke." It's a retrospective statement on the way things were versus the way things are now.
I would argue your point about the TFC models being distinguishable character models: In my eyes, they were not. They had different skins - the term in which I use in its old-school form, meaning textures - but they were not very distinguishable. Something Valve themselves have recognized publicly a number of times. So in that regard, I can see where he's coming from when he says 'just another shooter with army men'. Gameplay wise it was vastly different, but looks wise it didn't stand out as much as it could, and eventually did with Team Fortress 2.
Art direction has been more and more important as tech progresses. There use to NOT be dedicated environment artists. Level designers would build out AND dress their levels before, until graphic engines became advanced enough to where actual 3D artists needed to model & create artwork, as a team.
I'd say that goes a bit beyond just simple texture swaps. I dont think the game would have lasted as long as it did if the massive player base had a hard time telling classes apart.
What I said above is simply the impression I got while reading the article, which I thought came across fairly strongly. Much like art, though, it's certainly up to viewer/reader interpretation.
Ask most art directors/artists if they'd rather do a realistic shooter or a stylized shooter and most will say stylized... personally, I prefer to work on a title that has solid art direction either way. Just because your game is "stylized" doesn't mean it's more challenging to art direct or more fun to work on.
I'm sure the military is working on the friendly fire thing...