Close
Report Comment to a Moderator Our Moderators review all comments for abusive and offensive language, and ensure comments are from Verified Users only.
Please report a comment only if you feel it requires our urgent attention.
I understand, report it. Cancel

Epic's Tim Sweeney on the problem with photo-realism

Epic's Tim Sweeney on the problem with photo-realism

Thu 11 Jul 2013 11:37am GMT / 7:37am EDT / 4:37am PDT
PublishingDevelopment

Graphics will be "indistinguishable from reality" within 10 years, but that's where the real problems begin

Epic's Tim Sweeney believes that game graphics will achieve photo-realism within 10 years, though that endeavour merely highlights the myriad problems that raw power alone cannot solve.

Speaking at the Develop conference today, the founder of Epic Games marvelled at the million-fold increase in computing power for graphics since he first started programming. And at the current rate of progress, the environments in video games will be, "absolutely photo-realistic within the next 10 years."

"An indistinguishable from reality level of graphics," he said.

However, that will only offer the veneer of realism. Sweeney believes that creating truly convincing worlds requires solutions to a range of difficult problems.

"There are still a lot areas that will require ongoing research for probably the rest of our lives before we come close to approaching reality"

"That just moves the challenge of graphics to the problems we don't know how to solve. Like simulating human intelligence, animation, speech, lip-syncing. There are still a lot areas that will require ongoing research for probably the rest of our lives before we come close to approaching reality."

In a wide-ranging talk, Sweeney described in detail his personal journey to the summit of the industry. The need for rapid change was an ever-present message, and Sweeney believes that is more relevant now than ever before.

"It feels like we're going through the last 25 years of game history at the rate of four years every year," he said, addressing the huge changes wrought by the rise of agile, digital platforms for game creation and distribution.

These changes present a huge challenge to established companies, but Sweeney is confident that the culture of Epic Games will be able to withstand those forces and ultimately thrive - just as it did when the industry moved from 2D to 3D, and from PC to console.

"We've always recognised that the industry is in a state of constant change, and those that react fastest are the survivors," he said.

"We've always recognised that the industry is in a state of constant change, and those that react fastest are the survivors"

Indeed, Epic has already started to change in response to the new face of the industry. The Unreal Engine 4 has been built with a greater emphasis on flexibility and scalability than any other technology in the company's history.

"It means that now, increasingly, we can think about building one game and shipping it on every platform that's appropriate," he said, listing PC, console, tablet and the web as target platforms for Epic's future games.

"It means we can potentially reach a much larger audience, and it will be increasingly important worldwide. Consoles are specific to the Western markets - North America, Europe - but they really don't exist in Korea, China... I really think we can build one game that goes to all of those markets by supporting the appropriate platforms."

Sweeney said that there are already a number of projects of different sizes being developed with Unreal Engine 4 all over the world - "the rush is on right now" - and the "sweet-spot" for seeing the engine's capabilities will be, "the end of next year."

And the same is true for Epic, which has changed from the single-project focus of the Gears of War and Gears of War 2 years, to a new approach that supports the development of multiple games of radically different sizes - including the 30-man PC online game Fortnight, a larger unannounced AAA project, and another unannounced project being built by a team of three people.

15 Comments

Tim Carter Designer - Writer - Producer

574 317 0.6
Popular Comment
Nobody fucking cares.

See all those kids playing Minecraft? Do they care? That's the most photounrealistic game imaginable. A huge hit, too. Why? It's fun.

Posted:A year ago

#1

Eric Pallavicini Game Master, Kabam

292 193 0.7
See all those kids playing Minecraft? Do they care? That's the most photounrealistic game imaginable. A huge hit, too. Why? It's fun.
Well... Minecraft depth and fun with photorealism ? If accessible on low/average specs devices, unless you are a retro-gaming or pixel art fanatic, you'll switch.

Edited 1 times. Last edit by Eric Pallavicini on 11th July 2013 5:35pm

Posted:A year ago

#2

Adam Campbell Associate Producer, Miniclip Ltd

1,179 967 0.8
I agree, except I think many of the issues he highlighted are problems now (even without true photo-realism).

The number of times we've seen impressive looking games, with impressive looking ideas only to find the animations are wooden, speech/lipsnyc is off or A.I isn't very 'radiant' and so on.

But these are on-going problems and we'll have to rise to the challenge when it comes to solving them. All exciting areas actually and hopefully we'll see good progress in the coming generation of technology.

Edited 2 times. Last edit by Adam Campbell on 11th July 2013 5:52pm

Posted:A year ago

#3

Eric Pallavicini Game Master, Kabam

292 193 0.7
The number of times we've seen impressive looking games, with impressive looking ideas only to find the animations are wooden, speech/lipsnyc is off or A.I isn't very 'radiant' and so on.
Clearly, but because now it is all for show. When we'll reach the point where display, graphics, animation have matured, we'll eventually start working on real AI. But for now, if you want to sell an improvement to an investor that as no clue about what you are doing, it should better be visible. Unless we are in times of necessity, all that we care about is appearances. When we'll reach that point where our biological structure cannot make the difference between a billion and a trillion pixel resolution anymore (you get the point) then we'll work on something else that our intellect can perceive as a "visible" evolution, and more complex physics and AI are most probably the next in line.

Edited 4 times. Last edit by Eric Pallavicini on 11th July 2013 7:40pm

Posted:A year ago

#4

Ruben Monteiro Engineer

79 194 2.5
He already said this in 2009. Better dump the deadline for "sometime in the future".

Posted:A year ago

#5

Diego Santos Lećo Creative Director, GameBlox Interactive

25 26 1.0
More graphic power doesn't mean photorealism to everyone. I for example, love the smaller optimization burden. The new consoles will allow great "Indie graphics" with much less technical burden.

You don't have to worry about full screen effects dropping your frame rates, or taking details out of support characters just to keep the game at 30fps, etc.

Posted:A year ago

#6

Adam Campbell Associate Producer, Miniclip Ltd

1,179 967 0.8
Clearly, but because now it is all for show. When we'll reach the point where display, graphics, animation have matured, we'll eventually start working on real AI. But for now, if you want to sell an improvement to an investor that as no clue about what you are doing, it should better be visible. Unless we are in times of necessity, all that we care about is appearances. When we'll reach that point where our biological structure cannot make the difference between a billion and a trillion pixel resolution anymore (you get the point) then we'll work on something else that our intellect can perceive as a "visible" evolution, and more complex physics and AI are most probably the next in line.
Not true of everyone...

A lot of gamers aren't even aware of what we can achieve in these areas and the above are common complaints in games. Though when advancements are made they're appreciated.

To an extent, graphical improvements are mattering less than before, and perhaps its already time the focus should be shifting a little more. I wouldn't disagree that priorities (outside a few studios) lie elsewhere on the whole.

Posted:A year ago

#7

Andrew Goodchild Studying development, Train2Game

1,254 420 0.3
Well... Minecraft depth and fun with photorealism ? If accessible on low/average specs devices, unless you are a retro-gaming or pixel art fanatic, you'll switch.
However, whereas Minecraft was originally developed by one guy, and added a still quite small team when numbers sold wracked up, and therefore was amazingly profitable, with a photo-realistic version we'll be reading how the 6 million copies missed sales projections, it needed 9 million to break even, the company went bust, and they can't see where it went wrong.

Posted:A year ago

#8

Greg Wilcox Creator, Destroy All Fanboys!

2,193 1,170 0.5
Popular Comment
Oh gebus crisp... The BEST looking visuals in the world are meaningless when you'll still have doors that can't be opened to see what's behind them, characters getting hung up on geometry because they can't navigate OVER or around obstacles or move in the full range of motions actual humans do (mistakes included). Cut scenes and cinematic spots will look incredible, but in terms of gameplay, unless you have characters that act and react EXACTLY like humans (no canned animations), there will be people not convinced...

Posted:A year ago

#9

Eric Pallavicini Game Master, Kabam

292 193 0.7
To an extent, graphical improvements are mattering less than before, and perhaps its already time the focus should be shifting a little more. I wouldn't disagree that priorities (outside a few studios) lie elsewhere on the whole
Which is probably the reason why Indies are doing so well at the moment.

It is not about visual for them (they're back in the condition Andrew mentionned, small teams, limited means - both favoring innovation, the first making it easier to implement the second presenting challenges that needs to be overcomed with wits - but if the game sells in such an expanded market... wooohooo jackpot!!! and that is also why their prices are usually lower), it is about gaming culture and fun gameplay...

Edited 1 times. Last edit by Eric Pallavicini on 12th July 2013 9:00am

Posted:A year ago

#10

Shane Sweeney Academic

392 400 1.0
Popular Comment
@Tim Carter. Chill with the attitude. I think you might be in the wrong place.
I don't think Tim Sweeney is saying Photo Realism is the end game of video games, he is saying it will just be naturally achieved.

I can't imagine the source of your rage? If anything you should be championing Tim Sweeney's work. While his peers like Carmack were pushing for higher fidelity, and solving technically complex problems Sweeney was interested in making video game engines easier to use.

The Unreal engine has never been the most technologically superior engine. Cryteks fidelity is higher, Carmacks stuff introduced features the world had never seen (at the cost of compatibility and reinventing the wheel). Sweeney's focus since the very beginning was backwards compatible Middle ware, easy to use tools, and reducing the bottle neck on programmers even at the cost of graphics. This has been the back bone driving force behind the Unreal engine and has made it the #1 platform. The highest detailed engines have already lost the battle, Sweeney's vision of easy to use middleware has won.

You're a designer, go and try and design with the Minecraft engine? Wait, you can't. Unreal can do what you want, when you want and soon on what platform you want. Besides maybe Unity (which is clearly inspired by Sweeney's entire ethos of Engine building) is their a more user friendly middle ware? Certainly not one with the scale, scope or flexibility of the Unreal Engine.

This man is your champion, not your enemy. Chill.

Edited 1 times. Last edit by Shane Sweeney on 12th July 2013 9:35am

Posted:A year ago

#11

Benjamin Crause Supervisor Central Support, Nintendo of Europe

82 38 0.5
There are dozesn of games that are proof that the visuals are not the most important aspect of a game. Its the content and fun. If it looks good thats even better. But a game that looks great but which content is shallow is waste of time and money.

Posted:A year ago

#12

Saehoon Lee Founder & CEO, Pixellore

60 41 0.7
Ofcourse. Just having a great graphics doesnt make a good game. But using unreal doesnt mean the game play will be bad. It just mean that there is room for having a decent graphics if you want. And that cant be a bad thing. Unreal is just a tool so how you use it is up to the developers. It goes the same to the game play.

Posted:A year ago

#13

Andreia Quinta Creative & People Photographer, Studio52 London

224 590 2.6
Same issues as discussed in the comments of the FPS article. Graphics is 'easy' to get there. A.I., animations and interactions (doors, blowing stuff up) is going to be the real - decades long - brain puzzle.

Posted:A year ago

#14

Kareem Merhej Designer, infoLink-inc

21 27 1.3
He must literally mean for a photograph. Realistic graphics in a photograph in 10 years, sure.

There's just no way we have eyes/teeth/tongue/mouth/sweating/breathing/animation/multi-man-collision that all work together "just like real life" in 10 years in a game the size of GTA with all its city simulation and physics and AI.

Obviously I respect Tim (a lot) but this is just dreamspeak.

Posted:A year ago

#15

Login or register to post

Take part in the GamesIndustry community

Register now