 |
 |
|
If you enjoy reading this site, you might also want to check out these Think Services sites:
 Game Career Guide (for student game developers.)
 Indie Games (for independent game players/developers.)
 Finger Gaming (news, reviews, and analysis on iPhone and iPod Touch games.)
 GamerBytes (for the latest console digital download news.)
 Worlds In Motion (discussing the business of online worlds.)
 Game Set Watch (the Group's alt.game weblog.) |
|
|
|
| |
Epic's Sweeney: Games Are 'Factor Of 1000' Off From Graphical Realism
by Staff, Benj Edwards
|
|
| |
|
May 26, 2009
|
| |
Talking as part of a special Gamasutra interview spanning his career, Epic founder Tim Sweeney has suggested that "we're only about a factor of a thousand" away from perfect graphical realism in games, but it might only be 10-15 years time before it's accomplished.
Sweeney, who founded Epic Games as a PC shareware company in 1991 and is renowned as one of the game industry's all-time technical gurus, continues to play a pivotal technical role at the Unreal Engine and Gears Of War developer.
The comments were made as part of a major Gamasutra interview examining his career, which largely focuses on his company's fascinating early shareware days as Epic MegaGames.
However, the end of the interview also discusses Sweeney's thoughts on current technology, and Gamasutra is reprinting that section in full in this news story, as follows:
Does it scare you when you see Crysis... or, I don't know. What's the biggest competitor in the game engine market? Is it id's stuff, or is it another company?
Today, it's Crysis. Because Crysis is doing some things on high-end PCs that we're not doing ourselves. That's a tricky case though, because we could do vastly more than we're currently doing if we focused on supporting dual high-end video cards, which have about 10 times the graphics horsepower of a console today.
The thing is, that market is about 2% the size of the overall next-generation game market -- PS3, Xbox 360, and mainstream PC. So there's a real hard business decision: if you go the route that Crytek goes, you can beat us in certain areas in graphics, but you're really sacrificing the larger market.
Because you can't run that engine on the PS3 and that sort of thing.
Yeah. And some of the things they're doing are just conceptually incompatible with that level of performance. Wherever you have an order of magnitude performance difference, you can't really scale.
We can scale down in performance by a factor of three by going to a low resolution, dropping some textures, and things like that. But to scale by a factor of 10 -- you can't design a game with 10 times the detail and then scale it back to something that looks decent on the consoles.
You'd end up looking much worse than a console game that was just designed for the console specs. So they have real scalability difficulties there.
On another note, LittleBigPlanet reminds me of [Sweeney's early '90s shareware tool] ZZT, because you can build your own levels, and people do emergent techniques with a simple set of tools. Have you played LittleBigPlanet? Do you like it?
Yeah, it's brilliant. If I were creating a game with a really small team so that things like Unreal were ruled out, it's exactly the kind of game I'd want to create -- a side-scroller with really cool graphics using 3D and really cool physics -- and design it around the user community building stuff. It's a really brilliant project.
I think it's indicative of the direction in which that kind of game could go -- a sandbox game like The Sims or anything else. You can really go a long way with the simulation without creating complicated tools. It's not like there's some crazy Python scripting language in there; it's all done with really simple graphical stuff. It's a great idea.
Looking ahead, how long do you think it will be before real-time computer graphics are 100% realistic like a movie?
There are two parts to the graphical problem. Number one, there are all those problems that are just a matter of brute force computing power: so completely realistic lighting with real-time radiosity, perfectly anti-aliased graphics, and movie-quality static scenes and motion.
We're only about a factor of a thousand off from achieving all that in real-time without sacrifices. So we'll certainly see that happen in our lifetimes; it's just a result of Moore's Law. Probably 10-15 years for that stuff, which isn't far at all. Which is scary -- we'll be able to saturate our visual systems with realistic graphics at that point.
But there's another problem in graphics that's not as easily solvable. It's anything that requires simulating human intelligence or behavior: animation, character movement, interaction with characters, and conversations with characters. They're really cheesy in games now.
A state-of-the-art game like the latest Half-Life expansion from Valve, Gears of War, or Bungie's stuff is extraordinarily unrealistic compared to a human actor in a human movie, just because of the really fine nuances of human behavior.
We simulate character facial animation using tens of bones and facial controls, but in the body, you have thousands. It turns out we've evolved to recognize those things with extraordinary detail, so we're far short of being able to simulate that.
And unfortunately, all of that's not just a matter of computational power, because if we had infinitely fast computers now, we still wouldn't be able to solve that, because we just don't have the algorithms; we don't know how the brain works or how to simulate it.
So you'd have to create a perfectly realistic virtual human first to have perfectly realistic graphics.
Yeah, you'd have to simulate the brain and nervous system in the computer.
And circulation and everything. But that's probably going to be possible some day, don't you think?
Some day, yeah. But there's no Moore's Law for that stuff, and progress is very non-linear. Somebody must have a clear understanding of how a neuron works now and how it transmits to adjacent neurons, but they have no idea how a billion neurons combine together to create a brain and what parts of our brain are basically hard-coded by evolution, and which parts are based on learning, and so on.
And if you could simulate it all, how could you train it to be realistic like a human? Those problems are probably decades away from being solved. Those are things that may not occur in our lifetimes.
Just like perfect computer speech recognition: if you look at speech recognition, it's only gotten slightly better in the past decade, just by a factor of several hundred increases in computing power. That shows that those problems are not falling to brute force.
They're not following Moore's Law acceleration. Have you ever read any Ray Kurzweil's futurism stuff?
Some of it. Definitely anything like that is long-term future. It's just because we lack the algorithm and knowledge of how to model that, even given unlimited computing power.
Do you think that if the graphics were perfectly realistic that you would create a "final" version of the Unreal Engine? As in, you'd never have to make another one because you got it perfect?
I don't think we'd ever be done. If you look at special effects or anything like that, even if you have perfect rendering, you might have the inner loop of the renderer which processes sub-pixel triangle primitives and generates perfectly anti-aliased graphics -- that might be completely done, but still, the tools and the functionality that lets the artist create the environment is infinitely improvable.
Currently, our approaches are somewhat primitive if you look at it. If you want to build a city for a massively multiplayer game, the artists model a single column or wall, then they copy and paste it throughout the world. You'd run really large-scale procedural tools for defining road systems and defining buildings, and the structure and architecture of buildings.
You could really do a lot more to improve productivity than any current engine or modeling tool is doing. So I see decades and decades of work there, just to make ever-better tools. And special effects as well. Particle systems are only partly limited by computing power.
The other limitation is the tools that the artists have to control what's really going to happen, and to do that productively. If you can make tools more productive so an artist can do more in an hour, then that's an improvement too.
[The full Gamasutra interview with Tim Sweeney, including a mass of fascinating information on his shareware beginnings, is now available to read.]
|
| |
|
|
Someone equipped with fully-realistic rendering would not have to depict in the game the boring stuff you could see outside your office window--unless doing so makes sense within the context of what the game is trying to portray. Just like a film or a photograph, and just like existing games. You could already render a boring alleyway with an enormous amount of fidelity in a game. Just because you can do that doesn't mean that's what your game has to be about.
Having perfectly realistic rendering would not preclude interesting composition, interesting lighting, interesting movement, or any of the million other factors that make art capable of doing more than simply reproducing the real world with no artistic filter. To say that designers and artists would lose their capacity for creativity if they could depict reality in an extraordinarily convincing way seems to demonstrate an extremely low confidence in the creative talent in this industry. It could be argued that is justified to some extent, but I'd like to think people could still think of interesting things to show and offer for play.
From pure abstract to pure simulation. For my definition, pure simulation and realism is wenn we reach startrek holodeck simulation technology. So to me all games are first of games. Because a simulation is a simulation and not a game. But some games have more focus on realism and simulation, the grey position. So all games fall somewhere between pure abstract and Pure realism.
Because some realism things, don't make games fun or make them unbalanced.
Ofcourse there are a lot of people who are more gaming for escaping realism in games. But a lot others want to do things in games they wouldn't do in reallife. Like exciting cool job's who are dangerous, or doing only the fun part for a day. Job's Like a fighter pilot, police man, Soldier, criminal. Or job's who are out of reach.
Example is also sport games.
So I Play Little big planet, nice and fun. If only to play with friends and family. To kiddy for me to play on my own. Not my thing. But splitscreen Coöp.
But sometimes I want to play like a Deltaforce or SF recon, simulation, chalange.
Sometime i just want to shoot. Onrail shooters on the Wii.
sometime I just want to wreck things in a shooter, Nice that Physics get more attention.
That's why there almost always comes a realism mod, also for Unreal games. I think most of the gamer are more somewhere in the middle. COD4MW is such middle ground game.
Lookin out for ARMA 2 and OFP2 and AA3.
Let's say you have a dragon in your game. Would you rather it be displayed with 10 triangles or 20? With 50 triangles or 100? With a million or with a billion? The more detailed fidelity you can produce, the more exactly you can match your vision of what you want your dragon to look like. Want modeled scales instead of textured ones? It's possible at a high enough level of realism. Want finely-detailed stretch and squash like a cartoon or stop-motion animated character? Ultra-high realism will help.
Given that, I think that reaching visual perfection (if possible) is a great goal, and the sooner the better. I would love to be able to design my game and pass to the visual artists all of my fantastic concepts with confidence, knowing that anything I can describe, they can make. Sounds better than fussing over how much texture detail to put on the dragon's wings. And then, once the ultra-graphics are here and everyone has gotten used to them, we can get off the visual treadmill to some extent. We can stop debating whether games are more about flashy graphics or immersive gameplay because *everything* will have all the graphics you could want.
10-15 years? Can't wait.
Plus the ability for added realism only helps games that have a different art style.
I realize the overall point you're attempting to make, but I do think you have to reconsider this statement, when you factor in things like the fashion industry, or the movie industry. It's all "realistic" as we use real people for models and actors, but there is a whole lot of artistic design slapped onto them.
Oh jeez, Chris Remo already covered this...ah well I'll just leave it in.
To claim that reaching realistic visuals is a dead end is an incorrect statement because it ignores what games ultimately are, a set of rules and goals. You can have the simplest 2D visuals to represent the rules and goals, or the fanciest holographic visuals, or even a deck of cards. With a deck of cards, we're still seeing new card games. Imagination isn't restricted by the medium.
I did want to remind that simulation games such as sports games and racing games are a part of this industry and that they rely on realistic visuals (and sound and so forth). Games based on movie and TV IPs are also big business.
Here's a simple example I like to use to illustrate the coming complications of advanced graphics:
You walk through a city which is under attack, and you find a mount (a creature that can be used as a transportation method, like a horse). The whole city is under attack by boulder-throwing catapults and the mount is just sitting there as buildings around it are falling apart. Since you have a high graphical quality, life-like rendering and so on, you would expect the mounts to act like it's alive. So the programmers need to code AI for something that used to require none, so that the mounts can react even when the player is not riding them.
Now if a giant boulder lands next to it, it should react. Well that implies contextual animations. Already it's getting a bit complex, and the part of the city you're in is quite narrow, so animations must take the tight environment into consideration. What happens if you shoot a flaming arrow next to the mount's head? Does it attack you? Does it know what the arrow implies? What if a player is riding the mount, and someone shoots an arrow in its tail? Does the player lose control as the mount charges you or as it runs away, or does it just play a "hurt" animation and sound effect and keep standing there?
This issue is actually something we will have to face head-on very soon, but I believe that the studios who will have the most success will be those that understand that not all problems need to be tackled at the same time, and that games must still be released according to their respective schedules and make money. Epic used a very common-sense approach when they made Gears of War, and it paid off. I'm sure they'll still have the wit to take such a well-thought-out approach when the next generation comes to deliver a great game and actually make money on top:)
How can can we be 10-15 years away from perfect graphical realism in games if today we cant even achieve perfect graphical realism in films. This is with supercomputers chugging away for minutes to produce just one frame.
theres an pdf from nvidia 2003, elegance of brute force
now they come to the result performance needs to improve 36,000x to get close to human vision (perhaps thats close to 1,000-3,000 in todays numbers)
but they dont even mention the 600pound gorilla, more accurate lighting models eg GI, reflection etc
In addition, I'll offer the view that many people do not want so-called "realism" in their artistic entertainment regardless of the medium being used. Chris mentions photography and film as "photorealistic by definition" but actually, they are not, or at least not necessarily. There are lots of tricks that are done to film - both single frame and multiple frame - in order to convey a specific visual to the final audience. These tricks might be done during the original filming by using certain types of lighting or specific types of film, changing film speeds, etc, or they might be done during processing/developing of the film by using different techniques in order to generate certain traits in the final product.
Likewise, artstyles in Japan and Korea, or various artstyles throughout history, choose specific aesthetics rather than attempting any sort of "realism" for the final product. Some audiences might want photorealism, but many do not. We even have cel-shading used to make CG appear to be handrawn cels rather than photorealistic CG.
After all, there are good reasons why older games continue to be popular through re-releases in collections or on handhelds, or through the used game market. Frankly, the modern focus on "realism" is a big reason why modern games are frequently not as fun as older games despite anything else they may offer. Spend more time on story writing, character development, and design balance/pacing and less on visual hype and the property's profitability will last much longer.
* Sports games: The primary sales driver for the annually updated PES, FIFA, Madden etc is incremental graphics improvements. Gameplay is usually tweaked only and therefore secondary. Their steady march towards photorealism is the genre's commercial hearbeat and with the first two examples selling c. 15m-20m units each and every year, it is clear that the mass market games player agrees.
* Racing games: Likewise, this is a genre that thrives on graphical improvements and again is a substantial part of the market by value.
* Any historical games: Games like the excellent Total War series and indeed many of the 2nd world war games (e.g. CoH) do not seek to provide a stylised look and feel as this can compromise the historicity of their product. Photorealism allows for the potential of total historical authenticity.
* Simulations: By their very nature, these seek to simulate reality. The flight dynamic of flight sims has been mastered for well over a decade. What drives what is left of the flight sim market forward is graphical realism. A flight sim where you can fly around a photorealistic London, Paris or Tokyo would be more fun than one where you have simple, flat textured cities.
Escapism should not be confined to aesthetics. GTAIV's appeal is largely based around performing outrageous acts within a realistic faximile of New York. Reduce the graphics to a selection of blocks and take away the design proximity to New York and you are left with a game that will undoubtedly be fun but which is nowhere near as appealing as the existing version.
Kale - "If video games were to ever look flawlessly realistic, there's no longer any artistic value."
I would have to strongly disagree with you there, and here's why. The value of art is one of the highest within the games industry, it can lend itself to being one of the best affordances we can offer. I can't see the creative minds of today and tomorrow creating realistic games without any sort of gameplay behind them, this is where the synergy comes in. The difference between life and virtual life are and always will be (for the time being) the small intricacies that define us.
We copy people and animals. We study, refine and engineer ways of doing things so why should creating a 'realistic' 'game' be any different, it is a vital part of human nature.
Do you admire Rembrandt for his realistic work or prefer a more abstract Kandinsky so your mind can wonder in it's own 'virtual world', which is after all what our minds are - cages that confine our creativity and ideas.
15+ years spent on getting more sophisticated graphics tech, because suddenly the industry leaders were engineers maintained by graphic cards selling mafia.
We could rival theatre and books by now regarding story-telling depth and complexity of interaction, but no, they had to pimp the graphics up, those "engineers"
Thank you, ID and EPIC for being money whores.
Please avoid cinema comparison, it's completely out of place.
Why bash an improvement? Either utilize the advanced technology or go a different route?
This seems no different then fanboys complaining on forums about how X game is cooler then Y because they "SAY SO."
To make improvements in the superficial aspects, many forgot to improve non-visual(non-profitable?) ones, not only, but the first suffocated the last, we've seen a sheer regression in the gameplay, many games have life-like visuals but their gameplays are more primitive than frogger.
you dislike extremism, i dislike moderation when it's another word for lying. Even if things were more complicated than how i put 'em, why is it wrong to make 'em simpler? The message is simply to cure substance and only AFTER try to make the thing look nice. It's simple and extreme, but how can you say it's wrong?
Because realistic CG has robbed movies of all artistic value? Come to think of it, those have always looked pretty realistic.
Frightening that so many people agree with something so wrong.
(what hasn't happened with Oblvion, infact, which is the perfect example of technical sophistication with zero soul, an expensive benchmark, a safe source of money from Card makers)
So forget this argument, it leads nowhere, it's more important to remember that great graphics always stifle gameplay depth, with their malleability issues, with their costs, with their bugs, with the time needed to make 'em work and lastly with the fact that "good graphics per se always compensate lack of interactivity"
I know it's something too "righteous" to say, i know in the internet evil is glamorous, morality is sort of awkward, isn't it?
Apart from that - what kind of a budget would you need to create photorealistic content? I think that is unrealistic, too.
I also take issue with a premise that Con Quests made - that substance is always more important than appearance. In particular, I reject that the two can be separated from one another so out-of-hand. In a physical simulation, the detailed graphics are often the minimum feedback required to portray the state of that simulation. In addition, it's important to remember that just because we can come up with more accurate rules to govern the behaviour and appearance of objects, _that doesn't mean we always have to follow them._ Imagine a child's shovel-and-pail, covered in sand at a beach. Now make the pail invisible and see the shape of the sand scattered on its surface. In reality, doing this is impossible. In a digital simulation, it could be a minigame.