The Video Games Of The Future Could Look This Awesome

The Video Games Of The Future Could Look This Awesome

A new console generation is almost upon us, not to mention a new era in PC gaming fidelity. We know better graphics are coming, but just what do those better graphics actually look like?

They’ll probably look a lot like this.

This technology is called Separable Subsurface Scattering. It’s the product of “hours and hours of research, desperation, excitement, happiness, pride, sadness and extreme dedication” on the part of researchers Jorge Jimenez and Diego Gutierrez, and it’s designed specifically to be used for human skin in video games.

Not trailers, not movies, not cartoons, but video games. SSS has been written entirely in DirectX 10, and everything you see is being rendered in real-time. Don’t believe them, you can download the original demo here and try it yourself.

If you watch the clip and find yourself a little underwhelmed, that might be because it was designed to run fullscreen at 1080p. So try it out as nature intended for the full effect.

THE DAY HAS COME [Jorge Jiminez, via Rock, Paper, Shotgun]


    • its not a picture there are real time lighting effects and the camera revolves around the subject.

      One step at a time these guys only create the technology, it’s up to the developers to take it and rig it for facial animation hair etc and finaly animate it.

  • Looks awesome.

    But you’re gonna need really good quality facial animation to go with that though. Otherwise it’s gonna be super-uncanny-valley.

    Also, notice they didn’t have his eyes open. The dead eye problem with video game graphics is way more important to fix than the skin texture.

    • While I can’t necessarily say if it’s the case of this implementation, but the really uncanny look of eyes is caused by how they have multiple layers, and a lot of depth. Given the way subsurface scattering works, you could theoretically write a shader to handle eyes based off the same basic stuff, that would help.

  • I can guarantee graphics won’t look like this next gen. That model is ridiculously high poly and the texture is ridiculously high resolution. I don’t actually see anything in that video that is new or groundbreaking. Of course you can render a super high poly model with a super high resolution texture in an empty space with modern graphics cards. You’re not going to be able to render 10 different people with a suitably realistic environment to match them though.

    • Can you, Anon? Can you really? Well there you have it, ladies and gents… Don’t know why the discussion continued after this comment. Anon’s cleared it all up for us right there… With a guarantee, no less! Whoa! With that sort of awesome foresight power at your disposal, I’m amazed you choose to remain anonymous, sir. Think of all the good you could be doing! Anything else you can guarantee for us?

      • Yes I can. Current high end PCs aren’t capable of rendering scenes with this level of fidelity at anything approaching a playable frame rate, and as always the consoles will be using mid range PC parts at launch because they have to keep costs down, and it’s a given that there will be a new console from Microsoft and or Sony within the next 3 years.

  • When I look at this, and compare it to the faces in current games, I wonder why people keep saying we don’t need another generation of hardware yet. I was ready a few years ago.

  • We don’t need more face tech, what we need is improvements in clothing. Look at Mass Effect from 2007, the faces then were amazing, but the clothing hasn’t improved at all since then. Come back to be when you get clothing flowing properly, not poping out, contouring to the model correctly, and simulating real cloth correctly (ie not looking like plastic).

  • Realistic Graphics are nice and all, but give me some stylised artistic flare to show some talent of the artists rather than going from the gritty.

    • Bit of a mac fan boy? Move along nothing for you here son.
      Can’t wait for more dx10/dx11 stuff this is going to be awesome

      • Nope, quite the opposite, I hate apple (namely, iPods, iPads, iPhones, and Steve Jobs) with a deep loathing passion.

        I just know that direct x is a cancerous attempt to become a gaming and graphics monopoly.
        “Hey guys, check out how amazing direct x is! It’s so great, it does great graphics and doesn’t afraid of anything! … *cough*even though opengl has, for a long time, been able to do everything direct x can do!*cough* Quick everyone, go code your own direct x game for microsoft machines! Don’t forget, if you make xbox games, you have to give us money! What’s that, you want to code on non microsoft products? Too fucking bad.”
        The sad thing is, the gaming industry are jumping on the direct x bandwagon not because of its ‘power’, but because of its popularity.

        I will always use opengl, because it is, in most aspects, vastly superior to direct x, and I can code for non microsoft machines. When you’re making a product that will reach the widest audience possible, you’d be stupid to use direct x.

        • Be that as it may, if you are an OpenGL coder you’re doing yourself a disservice by not watching and learning from this guys research, just because he chose DirectX doesn’t mean the same principles can’t be applied to OpenGL code..

          • For a massive company, maybe, but I don’t see the logic in doing something twice to get the same result.

  • I remember reasing Disney comment that while realism in a graphical sense is appreciated for objects and animals, they found with research that the human mind didn’t accept/couldn’t accept and/or liking too what they were being shown

    In a nutshell a couple of movies Disney/Pixar have made with humans have been deliberately made to have unrealistic attributes purely for the acceptance factor.

    Someone mentioned it before, Uncanny Valley –

    • I think it’s a normal reaction, a sort of backlash against getting hopes up 😛

      (At least, that’s what it is for me)

  • There comes a point where each new generation of graphical intensity requires such an investment of time and resources and money that it becomes an untenable process.

    I would have thought we were getting pretty close to that point.

    Also, really emphasises the growing divide between the AAA cutting edge and the ‘rest’

    You can pretty up lighting effects and improve texture resolutions all you want, but there are, as ben mentioned earlier, other aspects of animation that need to catch up sophistication-wise.

    Extensive leaps in use of motion-capture and incremental improvements in technology might aid this, the tech used in LA Noire might help, but in order to keep animations dynamic in nature, there’s still a lot of work to do before it gets it right.

    • The technology already exists and is frequently used in Hollywood. I wouldn’t be surprised to see partnerships between game publishers and movie studios becoming the norm. Many movie studios already have gaming arms like Warner Bros, Disney and Lucasarts.

  • This is great progress, however we are still a while away… while this demo is able to run in realtime, there is no environment, objects, or other character and art assets present. I assume in its current state it is very GPU intensive, whilst the texture work and all is quite amazing, applying it to a low poly model is going to make it look very average. im estimating about 90k polygons in this model…

    Might be feasible for a fighting game .

  • Surprised no one mentioned the lack of hair. No eyelashes, the eybrows looked completely flat (when they were in shot at all) and the shaved head.

    Until the hair is there, and the whole thing is moving, not that interested.

  • The only thing I could think about was the music whist watching this video.. I don’t know about you, but I was waiting for him to turn into some sort of giant transformer robot. But none the less, looks pretty good from where I’m sitting.

  • SSS is Subsurface scattering, so Separable Subsurface Scattering should be SSSS lol

    Not sure why all the hate though, they are showing off real-time SSS, nothing to do with hair and animations. It’s still going to work with lower poly models and smaller textures. It won’t look as perfect as the demo, but it’s going to improve how skin is rendered in games.

  • That’s great and all, but I don’t think a 6670 could render such high detail at 1080p comfortably. For consoles to run that, the detail would have to be held back greatly.

  • Don’t let the neigh sayer discourage you…. This is the single thing that makes me proud to be a child of the video game revoulution of the 1980’s… This is solid proof in my eyes that video gaming is an art and that interactive media is the future of social entertainment… Congratulations!!!

Show more comments

Log in to comment on this story!