Whenever you got into gaming, we’re sure you were impressed by it at the time. It can be tough to remember, but over the last 30 years, we’ve moved from simple shapes floating around black screens pretending to be spaceships, all the way through simple sprites, full-screen cartoons, full-motion video and early experiments in 3D, and into blisteringly detailed virtual worlds that we can explore at will. Right now, the likes of Crysis and GTA4 are the bleeding edge. Before long, they’ll look bleeding awful. Our jaw-dropping first drive through the rebuilt Liberty City will be as quaint as Space Invaders to the gamers of the future as they sit in front of their PCs, munching sci-fi snacks such as Mars bars and Galaxies and Doctor Hula Hoops and wondering how we lived in the days before we could walk our game characters up to our own houses, peek through the virtual curtains and see ourselves sitting at our computers going “Neat”. Part of the problem with these games is that they set out to simulate reality, albeit in a stylised way. This is impressive at the time of release, but as time moves on, so do our techniques and technologies. In the case of 3D games, one particularly noticeable sign of age is that it took years before characters were capable of moving their lips while they talked, instead of just staring and nodding as a line of dialogue played. STILL GOT IT: The hand-drawn Broken Sword games have grown old gracefully Conversely, games that focus on artistry rather than technology often hold onto their looks surprisingly well decades later. Adventure games from the 1990s are a perfect example. Many look every bit as old and retro as they are, but no low resolutions or old processors can detract from the sheer artistry dripped into games such as Sam and Max Hit The Road , Gabriel Knight , or of course, Broken Sword , which was recently re-released on both Nintendo DS and Wii. Is it a bird? No, it’s a plane For most of the 1980s, artistic merit of any sort took a distant second place to simply trying to hammer the primitive graphics of the time into something that the player would find vaguely recognisable. Early PC developers barely bothered at all. It was far easier to draw a map and make a strategy game, throw together something simple such as an Arkanoid clone, or sidestep the thorny issue completely by developing yet another text adventure. The PC was still thought of as primarily a work machine and it would be years before that preconception changed. NEW ANGLE: Isometric games add a level of depth to early gamers’ adventures What separated this era of gaming from the likes of Crysis (if you ignore roughly 15 quantum leaps in various areas of technology) was that tricks such as these were essential. Elite used wireframe 3D graphics on most platforms because that was as much as they could hope to handle. Most games assumed the player would understand that principle, although some built it into the fiction of the game. Starglider was one of these, shipping with a novella to explain that the wireframe graphics were actually a tactical aid built into your HUD. Reading it was also your best way of realising that the game wasn’t actually set in the depths of space and that the reason you took damage if you flew too far in a downward direction was that the big black bit at the bottom of the pitch black background was in fact ‘the ground’. Games of this era were nothing short of a war between programmers and their machines. The Commodore 64, released in 1982, was a surprisingly powerful system, but games on platforms such as the ZX Spectrum could be instantly identified by their bright, clashing colours – particularly when dealing with conversions from more powerful arcade cabinets. The art style was mainly a way of compensating for the system and used tricks such as setting translucent characters against coloured backgrounds, or promoting the graphical details by making the active game area a single colour, with objects outlined in black and all other colours saved for the frame. BACK IN TIME: Space Quest IV used time-travel to let its hero revisit his own prequels Sprites were simple and often very stylised, Dizzy the adventuring egg being one of the most enduring icons. Most games were happy to throw almost anything into the mix, from remote-controlled heads to a Monty Python stomping foot. The most surprising thing is that some of the creators weren’t high. Pretty and programmed The PC was largely left out of all this craziness through most of the 80s, thanks to developers focusing on three rather serious genres: adventures, RPGs and simulators. Technology limitations meant that the early simulations had enough to worry about just drawing the ground and a couple of aeroplanes to shoot at, but the RPGs almost went out of their way to be ugly. Graphics were kept firmly in their place, and that place was frequently a small box framed by the far more important stats and menus. It wasn’t until Ultima VII in 1992 that we saw an original PC RPG where the visuals stood up as state of the art against its contemporaries. PRETTY FACE: Ultima was one of the first RPGs to make an effort to look attractive Adventure games took a totally different direction, to the point that, until 3D came along, they were the place to see the latest graphical styles. The first major graphic adventure, King’s Quest , released in 1984, was designed as a showpiece for the PC. No mean feat, considering how bad our favourite platform’s graphics were at the time. The basic colour graphics format was the hideous Color Graphics Adapter. This typically ran at a resolution of 320×200 and resulted in some of the ugliest graphics seen by mortal eyes. UGLY PALETTE: CGA graphics were barely passable in 1984, and time has not been kind It had 16 colours to play with, but they were split into two palettes containing six hues (with two blacks each), which developers had to choose from. The first was the ‘winter’ palette, offering cyan, white, grey/white and magenta. The second offered green, red, brown and yellow. There were ways of squeezing more power out of CGA, including setting the background colour to blue and thus having the full RGB set to play with, or switching palettes while drawing the screen. Another trick was to abuse the poor quality screens at the time, putting down dithered patterns of dots and letting the display blend them into the required colour. All very clever stuff, but clearly hack territory. Nobody at all shed a tear when CGA was booted aside in favour of EGA and VGA. To catch a sprite In the days of CGA, everything was incredibly simple. With EGA, artists could finally start cutting loose. Backgrounds became more detailed. Shading could finally be more subtle – if not by much – than simply overlaying dots on bits of the screen that were meant to be in shadow. However, in both cases, the results looked like the computer generated graphics that they were. Backgrounds were clearly made up with simple lines and fills, with blocky characters overlaid on top. This changed dramatically when VGA entered the picture. VGA allowed for 256 colours at once, giving artists much more freedom. Games such as King’s Quest V and Monkey Island 2 replaced computer generated backgrounds with scanned-in paintings, giving their worlds an incredible sense of detail and atmosphere. Spare colours could be used for effects, such as flickering lights, gradients and proper shading. DETAILS, DETAILS: VGA games used every trick going to make backgrounds look attractive Interestingly, this usually only applied to the backgrounds. It quickly became acceptable to have different styles for the overlaid characters, typically keeping them cartoony, regardless of their settings. This was primarily a logistical issue, given the number of animation frames that would have to be stored and created, but few minded. Many games were quite happy to chop and change media types on the fly, whether by kicking off with a 3D rendered introduction (as in King’s Quest VI , Syndicate and almost every early CD game), or showing detailed pictures of characters in conversations. One media mix that never worked well in this context was FMV, partly due to the often woeful production values of the games that used it, but mostly because of issues with converting footage into something that could be used as a game sprite. Low resolutions turned them into a blurry mess and higher resolutions hammered home how badly the non-antialiased sprites sat in the world. As just one example, an FMV character will always have more superfluous detail than a cartoon one, leading to more obvious problems if their walk cycles get broken – frequently making it harder to read body language. Rendered 3D figures suffered from these problems as well, but rarely to such a degree. The only games that made the leap successfully simply used the original footage as a starting point, rather than trying to implement it into the game as-was. These included Prince of Persia with its rotoscoped character animations, and Tex Murphy , which avoided having 2D versions of its characters in the 3D game world wherever possible, and simply cut to a full-on blue-screened FMV sequence for conversations and important actions. It was jarring, but consistent. All of this gave the early 90s a fascinating look. It was an era of mixed media, with classic animation sitting alongside revolutionary new ideas. Free of the constraints of flat colours and simple shapes, developers jumped at the chance to create unique worlds. NO FUN: We’d scream too if our world was made up of such clashing styles In adventures, we saw everything from the pure cartoon style of Day of the Tentacle to photographic/3D rendered hybrids in the likes of Lost in Time , to early attempts to put real people into the mix during the horrific age of interactive movies. One of the most bizarre was Darkseed , which featured two worlds: ours, and a Dark World based on the biomechanical paintings of HR Geiger. (There was talk at the time that the artists needed therapy due to this, but we call bull on that one…) Before there was 3D The success of 3D accelerator cards heralded the birth of a new era for PC games – one where 3D is mother and father and little sister all rolled into one. This is a good thing. Sprites have to be created frame by frame, which makes them inherently limited. Put a skeletal structure in a 3D model and it can do anything from pick up a laser gun to dance the tango, not to mention perform subtle actions such as making eye contact or following you around the room. 3D models also have depth, whereas sprites are just cardboard cutouts. Using the same engine for them and the backgrounds also helps reinforce the notion of a consistent world that characters can interact with. However, PCs powerful enough to make the most of 3D are a recent development and for a long time, sprites vs polygons was a genuine debate. Elite had wireframe polygons back in 1984, as did the first Star Wars space simulator, X-Wing , in 1993. Until Quake in 1996, which also gave the mainstream its first true 3D shooter, almost every FPS was sprite-based, with exceptions such as Terminator: Future Shock few and far between. PROGRESS: Floppy disks eventually became big enough for proper cartoon fun The main reason for this was that sprites offered more detail than the polygons of the time could handle. Much like the early days, it was fine to have simple iconic shapes for spaceships and the like, which is why X-Wing ’s instantly recognisable ship silhouettes worked so well. However, early games didn’t even have texture mapping (instead relying on simple colours and shading based on the position of the area’s light source) and 3D characters were awkward-looking, low-polygon affairs. With sprites, if you could draw it, the computer could handle it. When making Doom , id actually modelled several of its more complicated characters in clay then photographed them from multiple angles to maintain consistency. The first Wing Commander games used a clever trick to fake space, simply moving and scaling sprites in front of the camera to add depth. It fell down when approaching capital vessels, but made for instantly recognisable ships and graphical assets that sat comfortably alongside the character portraits used in menus and cut-scenes. SERIOUS FIREPOWER: As dodgy as it looks now, Outcast was a stunning game in 1999 – if your PC could handle it. Most couldn’t 3D was much slower to impress on anything other than a technological level. Quake was a particular disappointment, regardless of its technical prowess. Doom gave us a whole cast of memorable enemies to shoot at, ranging from the humble Imp to the awe-inspiring Cyberdemon. Few of Quake ’s blocky, blurry opponents came close, outside of the leaping Fiend and hulking demon Cthon. Not until Sin and Half-Life in 1998 did 3D characters start to impress and even then, the fact that most of them were still animated frame by frame made them feel stodgy. Half-Life ’s skeletal animations provided freedom, even if they did usher in several years of comedy ragdolls. Adventures in hyperreality As with the dawn of 2D, 3D games quickly began splitting art styles along genre lines. This was typically done for pragmatic reasons rather than stylistic ones. The more restrictive the game, the more realistic its world could be. Driving games such as Midtown Madness , Driver and GTA could get away with something that pretended to be a real city by confining players to roads and other external locations where real-world textures could turn a simple box into a credible store or tower block. EASY RIDE: Driving games have it easier than most. You don’t see much detail at 100MPH Most games stuck with a different approach: creating the illusion of reality, not trying to recreate the actual thing. The more games tried to do this, the more they hit the Uncanny Valley problem – deviations from the expected that break the suspension of disbelief. The simplest solution was to shunt things around a little by setting a game in the future, or in a lab, or in many cases, just not trying too hard. When Tomb Raider 3 set a level in Saint Paul’s Cathedral, the challenge was conveying that vibe with a game engine that had only just developed the ability to build levels using triangles instead of square blocks, not recreating reality. Similarly, Deus Ex took great pains to trash New York and Gabriel Knight III devoted as much effort to recreating the village of Rennes-le-Chateau as the Tomb Raider team did to King Midas’s crypt. It didn’t matter. Much as the classic 2D games hinted at a world beyond the bits that you got to visit, the best 3D games feel like there’s more going on behind the invisible walls that block your passage. TAKING OFF: Despite the dramatic music and statues, you can still cross Stormwind faster than you can say ‘Not very big, is it?’ It’d be wrong to say that current games aren’t constrained by technology, but the challenge is different. With modern graphics, we can create almost anything to an acceptable level. The challenge comes from the difficulty of creating all the necessary assets, with level design stretched across multiple disciplines, from 3D modelling to animation and physics calculation. The next step, along with increasing the polygon count and texture resolution, is likely to be to take as much of the load away from the artists as possible to create bigger worlds that retain the handcrafted look. Wherever games go next, they’re going to look fantastic. Related Stories Exclusive: Native iPhone 3G S gaming a long way off Wii bowling ball accessory breaks cover Steve Ballmer confirms Natal Xbox 360 for 2010 Alt-Delete: Hollywood does videogame comedy Activision threatens to stop making PS3 games
Read more:
In Depth: The evolution of gaming graphics
Download and watch part three of the Section 8 behind the scenes series. In this episode, the dev team talks about the unique dynamic combat mission feature and deployables system found in multiplayer.”
