Future Processing Power

[GalaxyS5]

When do you guys think we will have video cards with enough processing power to make games look exactly like real-life? What would the specs look like for these cards?

 

[defaultluser]

Definitely tomorrow.

You can already render small scenes that look pretty realistic. It's doing it for an entire game that gets dicey, and quite a bit more expensive to develop.

https://www.youtube.com/watch?v=sYX9I3ONHc4

 

[KazeoHin]

As video processing gets more and more powerful, and graphics get more an more realistic, our eyes become more and more attuned to the small details that are missing. Professional artists can spot a 3D render from a photo instantly if asked to, and video games are all about taking shortcuts and smoke and mirrors, so the trained eye can pick out those missing details even easier. We will never reach that threshold, yet we have reached it many times. Suspension of disbelief is the act of making our brains perceive what we see as 'real', even when the visuals are nowhere near real. When I played Turok on N64, I thought I was looking at something real, my memories of the game are photorealistic, nostalgia forces me to remember what my brain perceived, not what my eyes saw. Quake 2, UT2003, Far Cry, Doom 3, Crysis... All of these games had that effect on their players.

Yet all these games are obsolete, with bad graphical fidelity by today's standards.

 

[DeathFromBelow]

Its like asking when paintings are going to look more realistic. Realism isn't the point of games.

 

[TheOneKnownAsMe]

Games already take thousands of man hours due to the amount of work it takes for the artists to create such high-detail 3D assets. Quite frankly, we won't see that until there is an effective way to reduce the amount of work involved (For a single tree you have a 3D high poly mesh, multiple LODs, plus various complex materials that include whatever textures you need. Now, multiply that by every asset in the game.) Or until we find a way to streamline the process and make it more efficient.

However, with the pipelines in place today, that level of detail simply isn't viable in a game. It's possible, and I'm not worried about the hardware it will take to render it, it's simply the constraints placed on the actual creation of the game.

 

[Hornet]

We can already do that easily, just not in real time.

I don't think real time ray tracing is viable anytime soon, considering the hardware performance and the time it takes to render them for Hollywood movies at the moment. You'll need multiple folds of that performance to do it in real time, and cram all of that performance onto a single chip. Don't wait around for it.

Its like asking when paintings are going to look more realistic. Realism isn't the point of games.

Depends on what game, but in most game, the basic idea is to create a fictional world that the player could immerse themself in. You want to feel like you are actually there in this virtual world. Graphic fidelity is an important factor that enable this. For example, GTA is a fictional copy of our real world where you can be a thug. The more closely this fictional world resembles the real world visually, the more immersive the experience would be because the player will feel like they are in a real world (except that they can do things here that they cannot in real life), rather than a pixelated cartoon movie.

 

[LordEC911]

Don't forget the uncanny valley.
Once you get to near-photo realism things get "creepy."

 

[Liger88]

Don't forget the uncanny valley.
Once you get to near-photo realism things get "creepy."

A lot of developers these days also seem to be shunning graphically exotic games for something that feels more personal to them. Don't get me wrong graphics are pretty awesome when applied these days, but I am either getting too old or things have gone so far that I no longer get joy out of trying to find someone hidden in the foliage. I mean it gets to the point that the environment looks so good it's distracting and almost too realistic that you can't see people clearly anymore because they blend in so well.

I guess some people love that ultra type of realism. I for one immediately get disgruntled when I have to battle the vibrant surroundings to just do the basics of the game.

I imagine they'll keep pushing for photo realism, but like some developers who've already voiced their opinion, they don't think that's going to be what people want when it comes.

 

[DeathFromBelow]

Depends on what game, but in most game, the basic idea is to create a fictional world that the player could immerse themself in. You want to feel like you are actually there in this virtual world. Graphic fidelity is an important factor that enable this. For example, GTA is a fictional copy of our real world where you can be a thug. The more closely this fictional world resembles the real world visually, the more immersive the experience would be because the player will feel like they are in a real world (except that they can do things here that they cannot in real life), rather than a pixelated cartoon movie.

As was stated previously, there may be an 'uncanny valley' effect going on. I find games from the late 90s like Thief and Half Life to be very immersive. I notice everything that doesnt look quite right in newer games like Crysis. I'd rather see developers focus on good gameplay rather than endlessly chasing 'realistic' graphics.

 

[limitedaccess]

The issue with realistic graphics is not limited to computational power. Games are not static and this is where the issues come in.

http://www.gamasutra.com/view/feature/4035/from_the_past_to_the_future_tim_.php?print=1

Looking ahead, how long do you think it will be before real-time computer graphics are 100% realistic like a movie?

TS: 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.

As was stated previously, there may be an 'uncanny valley' effect going on. I find games from the late 90s like Thief and Half Life to be very immersive. I notice everything that doesnt look quite right in newer games like Crysis. I'd rather see developers focus on good gameplay rather than endlessly chasing 'realistic' graphics.

The "uncanny valley" is not a fully understood or explored issue at the moment. One theory regarding it is that the actual problem (aversion) is caused by flaws and differences in each aspect of the representation.

So let's say you have an in game model that is perfectly photorealistic while static but what happens when things are dynamic (games are after all dynamic and interactive)? The problem is we do not have the ability to feed a computer the data currently to actually accurately model a real person, so how he moves for instance would be extremely inaccurate.

This is what describes the issue you're facing. You're newer modern game, Crysis, tries to be photorealistic but cannot and has severe flaws in it's representation. While you're older game is not viewed as attempting to be photorealistic and therefore no aversion takes place.

You cannot fully motion capture and hand animate every scene in a game outside of cutscenes. So just having the resources currently available to big budget films for CG would not solve the issue.

 

[Quix]

I'd say that as long as we're using raster technology, no. Ray-tracing has a lot better potential to seem realistic, but we're no where near yet. To look truly "realistic" there are a lot of factors we're not even close to getting right. But saying that, I don't really care if my games look realistic or not.

 

[sk3tch]

As video processing gets more and more powerful, and graphics get more an more realistic, our eyes become more and more attuned to the small details that are missing. Professional artists can spot a 3D render from a photo instantly if asked to, and video games are all about taking shortcuts and smoke and mirrors, so the trained eye can pick out those missing details even easier. We will never reach that threshold, yet we have reached it many times. Suspension of disbelief is the act of making our brains perceive what we see as 'real', even when the visuals are nowhere near real. When I played Turok on N64, I thought I was looking at something real, my memories of the game are photorealistic, nostalgia forces me to remember what my brain perceived, not what my eyes saw. Quake 2, UT2003, Far Cry, Doom 3, Crysis... All of these games had that effect on their players.

Yet all these games are obsolete, with bad graphical fidelity by today's standards.

Well said.