Maximum Graphicsability: CryEngine Aids Industrial Design

Industry: friend to trees and plants and flowers, clearly.

Here is a very interesting thing that just so happens to have yielded intensely pretty results. The short version? We get to go “Ooooooooooooo high dynamic range spectral gazelle mapping slobber slobber moo” but also “Oh yes, hmmm, oh indeed, yes.” It’s all thanks to Enodo, a French company that’s realized games could have tremendous practical potential in the realm of real-world industrial design. Its plan, then, is to use CryEngine 3 to create interactive models of the building forests (and sometimes, regular forests) of tomorrrrrrrrrrrrow. Now, today’s demo is just a proof-of-concept video, but goodness, would you look at that?

Soon, we will be able to design the greatest elephant. That, I can only presume, was the goal of this little simulation. Or perhaps this:

“Enodo bridges two worlds, with only one purpose: to level up our client’s business by questioning cutting edge Real Time 3D technologies, to show that the mechanisms governing the world of video games are powerful and will revolutionize the way to understand urban or industrial projects, their development, and their communication.”

It is, admittedly, mired in painful business nostril-delivered buzzspeak, but I definitely like the idea. And honestly, this just seems like the tip of the iceberg. Imagine being able to tour an apartment or house you’re interested in even though it’s on the other side of the country, continent, or even planet. Or what if we had a way to suggest changes in city infrastructure with simulated conditions as evidence? “Yes, Mr President, I can indeed prove that turning the interstate highway into a giant, loopty-loop rollercoaster track will improve traffic flow 147 percent.”

Seriously, though, there are some fascinating game-related possibilities on the horizon, and they don’t involving shooting dudes or adding arbitrary point systems to real world tedium in a problematic attempt to up the “fun” factor. I definitely like the idea of games improving the world (beyond providing art and entertainment, because I do believe those improve the world as well), so I’m interested to see where this all goes. How about you?


  1. Drayk says:

    Well, lots of architecture programs already let you tour a 3D version of your future house.
    Sure it’s not the graphics you can achieve with Crytek 3 engine, but it still gives you a good sense of proportion and space.

    As for modeling road congestion, we already have models, again it’s not as entertaining or fun, but it works (more or less)

    I think uses for this kind of technology are strictly in the realms of luxury items.

    • elgonzo says:

      Well, those game engines have some appeal to certain architects and landscape designers. Well, to be more exact, their sandbox editors do. Aside from the low entry costs, the promise of these sandbox editors lies in being simple rapid prototyping tools during the zoning/programming steps (programming as in “architectural programming”, not game engine programming).

      • AngoraFish says:

        PC gaming and design tech have less overlap than many people assume. Existing design tech for the PC is at least five years ahead of PC gaming tech when it comes to simulations. There isn’t a lot of need for real time particle physics or ragdolls if you’re simulating a building or freeway.

        Having done full 3D flythroughs with shadows accurate to latitude, longitude and time of day in AutoCAD2010, the graphics aren’t half bad either.

        In fact, there’s very little in that video that can’t be done in SketchUp right now.

        • Faxmachinen says:

          If by “aren’t half bad” you mean “aren’t half bad for a house made of Lego”, then sure. Also, I don’t think the point of this thing is simulation, what with game engines not being able to simulate much of anything.

        • elgonzo says:

          If you speak about CG for product design vs. game graphics, i totally agree.
          Industrial design packages are all about geometric precision and a visual appearance that is as close to reality as technologically possible. Interactivity and high frame rates are of lesser concern.

          Game engines serve a different purpose. Naturally, they have to balance visual fidelity vs. frame rates (interactivity). Geometric precision is not really a requirement here.

          However, if you speak about landscaping, things change quite dramatically. Contemporary industrial design packages do a piss-poor job regarding visualization of (urban) environments. Which is a strength and purpose of game engines — obviously, since most 3D games require immersive 3D environments. Needless to say, the term “sandbox editor” hints exactly at that….

          • skittles says:

            I do so recken there is a market available here. Landscaping and urban design in something like CryEngine coupled with an Occulus Rift, definitely a big step forward for visualisation projects. The problem with existing visualisation projects is that so often the resulting product looks quite basic and means very little to the vast majority of users except those who originally created the visualisation. The vast majority of such projects the creators make a very simply ‘flythrough’ that means squat to most people looking at it. Making it interactive, particularly with technology such as the Rift would make it much more accessible. Just like what was in this video it would be excellent for urban planning, where quite often the user is not someone after an exacting plan. Although I do wonder what the African bits were meant to be about.

      • Quickpull says:

        I’d say there is no advantage that a game development tool offers over the industry specific design tools we have these day. Hell even SketchUp is a superior option for rapid prototyping of structures. In theory it could have some niche as a presentation tool, but even there I am skeptical. Generally when you need some nice looking presentation tools, you use things like Maya and 3D Studiomax and produce a pre-rendered video.

        So what would be the advantage of using a game engine? Real time exploration of a design sounds novel, but I cant imagine someone putting crysis version of a design in front of their board members or stock holders and saying “ok guys, look with the mouse and move with the keyboard”. At least not yet.

        Not to mention that using a video instead gives you a lot more editorial control about what people see to play up the things you think are important. It also lets you fake things to get really nice results without having to perfectly model everything.

        • elgonzo says:

          No, game engines (and more or less other realtime CG engines for that matter) are not meant to produce eye candy for pitch presentations. In this area, off-line rendering will still dominate for quite some time.

          Contemporary game-engine based tools are not meant for architects which deal with indoor projects or the construction of buildings. Game engines are designed around a compromise between visual fidelity and frame rate (interactivity). For any indoor architecture work, you will need geometric precision, and a somewhat accurate lighting — which might not be the easiest thing to achieve with tools based on a game-engine.

          However, game-engines (and their related tools) can fare much better in the area of urban planning and landscaping. Just with the sandbox tools provided with SDKs, amazing results can be achieved comparatively easy and quickly.

          Mind you, regarding landscaping / urban planning, geometric precision down to the millimeter is not necessary, and accuracy regarding lighting/reflections/refractions are less demanding than what you would need to do for indoor design projects.

          Especially if the project is about urban planning, having an interactive display (whether publicly, or just for the mayors office) might actually be much more impressive than some glossy printouts of some Maya renderings.

          Let me come back to the sandbox editor tools of the game engines. These sandbox editors allow you to evaluate the capabilities. Which is crucial if you are searching for a license-able 3D rendering pipeline.

          Also, in many cases, the standard sandbox editors from the SDKs will be sufficient for urban visualization.
          In a way, that’s what they were made for — maybe not exactly for urban visualization, but for creating environments which look compelling and in which you can wander around.

          Perhaps, those sandbox tools might not be suitable and fit for the task in all cases. But even then, the game engine will have the necessary APIs (which are also used by the original sandbox tools), which makes it possible to extend the editors or even create your own application. Basing your own applications on top of a 3D engine of an industrial design package is a license and support nightmare, as the business model of AutoDesk and alike is not really in licensing their 3D engines.

          Google Sketch-Up does not even enter the picture if you speak about scenarios where game engines can provide a benefit…

          EDIT: I forgot to mention scripting, which is supported by most game engines, and which is a substantial feature when creating believable environments. The video shows this quite neatly…

          • Quickpull says:

            Maya can do all those things with equal ease and better looking results. Again the only thing game engines can do better is run in real time because that is what they are made to do well. And that is of little use to a designer.

  2. Don Reba says:

    If 3D models of real-world architecture become easily accessible, there will be loads of controversy around people placing them into shooters.

    • Gap Gen says:

      I assume companies aren’t likely to release their stuff into the public domain.

    • SominiTheCommenter says:

      You mean, like every single FPS that takes place in landmark? It’s not exactly novel.

      • Don Reba says:

        I mean like that student who got expelled for playing Counter Strike in a model of his school.

        Regarding whether companies are likely to release their stuff into public domain — if you can tour an apartment or house online using client-side rendering, then you have the 3D data and can do other things with it.

    • calibypolege says:

      If you think Susan`s story is cool…, last pay-cheque my cousins girlfriend also broght in $4583 putting in a seventeen hour week from there apartment and the’re best friend’s step-mother`s neighbour has done this for six months and made more than $4583 in their spare time at there computer. use the guidelines here
      (Go to site and open “Home” for details)

  3. SuicideKing says:

    Oh look, noCrysis, since it’s so peaceful.

    And open world.

  4. Raiyan 1.0 says:

    Cry, Engine. Cry.

  5. Kein says:

    Every time I see an article like that only one question pop-ups in my head: WHY DON’T YOU USE SPECIAL TOOLS FOR SUCH KIND OF TASK. There are HUGE amount of professional programs made EXCLUSIVELY for such kind of work.


    • Ultra Superior says:

      What kind of authorized AutoDesk reseller are you?

    • Gap Gen says:

      I assume this is more for PR stuff than modelling per se, unless they’re building a new tool directly in CryEngine. Perhaps it’s easier to sell your proposed building if the graphics you have for the 3D mockup look prettier.

      • neolith says:

        It is – and that’s why this stuff is usually prerendered and not realtime. Also creating stuff for realtime is A LOT more work than for prerendered viz.

    • elgonzo says:

      There are a huge amount of (usable) software products for landscaping on the market ???

  6. MOKKA says:

    I want to level up my business as well.

    • Gap Gen says:

      – Staff of the Profit
      – Boots of Human Resources II
      – Fireball

      • deke913 says:

        You sir owe me a set of nostrils. My current set just had coffee shoot out of them.

        • Gap Gen says:

          Remember when you were asked to write your signature on the biscuit that came with your coffee? That was you signing our wafer. Waiver. Eh. Kinda works.

      • solidsquid says:

        Aquired Flaw: Bob in Accounting

        Bob in Accounting is a figure who poses fear wherever he treads. His refusal to use computers for any form of calculation, rejection of any budget applications which aren’t filed in triplicate by hand on an E532b form and constant complaints about how much worse things have become since the good old days will have an impact on your bottom line until you can find a way to shift him to another department

        10% chance of project not finishing on schedule
        5% drop in morale of accounting staff
        3% drop in morale of HR staff
        Gain the “Put Bob on the Case” ability to use on unwanted projects which office politics forced on you

      • Jamesworkshop says:

        It not all about “prophet”

    • SominiTheCommenter says:

      Just fill the damn RJ45, Jenkins!

  7. theodacourt says:

    I wonder if we will ever reach something like the matrix with this stuff and simulated worlds.

    • Cytrom says:

      We probably already had, and you lived your whole life in it without ever noticing it.

  8. Mr. Mister says:

    Bekenstein boundary bitches.

  9. Brayduck says:

    Am I the only one to whom this video felt… cartoonish?

  10. The First Door says:

    That is exceptionally pretty! Especially the detail on the flowers, very lovely. Still though, is there a term more nebulous that ‘real-time simulations’ when you don’t define what you are rendering on? For example, one of my favourite head tracking examples uses a 14 (now defunct) Cell Blades in a cluster and is ‘real-time’:

    link to

    • elgonzo says:

      The pretty flowers are… um… pretty.
      But if that would be the competitive highlight of Enodo’s products, that would be a not so pretty fail…

      It is meaningless to extrapolate the “quality” of realtime just from the amount of processing nodes. It ignores crucial parameters about how many viewports (view channels, stereo/mono) are used, what are their resolution(s), what kind of user input technologies are being utilized (what kind of tracking systems), etc…, etc…

  11. warthog2k says:

    Didn’t the Unreal engine also get used for similar architectural prototyping / demo-ing a few years back? Quick google hasn’t unearthed anything useful, but it’s ringing bells in my head…

  12. Wang Tang says:

    Random rant: FCK this GEMA shit.

    For anyone else from Germany, here’s the demo non-GEMA-fied: link to

    • stahlwerk says:

      But think of the composer! How should he feed his family if not for the measly generous amount the Archbishop Colloredo GEMA pays him probably never every month.

    • Zaarin says:

      Thanks a lot for the alternative link. GEMA really makes me angry at times…

    • elgonzo says:

      GEMA: The never-ending pursuit to protect German people from evil rock/pop music.

  13. mezron says:

    I remember the same thing was done with the Unreal engine, and Quake 2 also in the late 90’s. I don’t know if it ever went anywhere, but I thought it was interesting at the time
    link to
    link to

  14. CantankerousDave says:

    There’s also that project by Titanic obsessives to do a full digital recreation in CryEngine3.

  15. alkonaut says:

    This makes so much sense. For realtime rendering professional apps have no chance in competing with game engines, especially in areas like procedural foliage, character behavior etc, and also when it comes to squeezing every fps available from a wide range of hardware. If I created frostbites/unreal/cryengines I’d make sure to tap into that bit of revenue by licensing to Dassault/Autodesk and selling them tools to bake the game worlds from architectural models.

  16. Engonge says:

    what is this sorcery? Where is the shooting?

  17. Ny24 says:

    Why are they sitting like that? It freaks me out …

  18. Thiefsie says:

    As an architect, I must say this has scant chance of actually becoming mainstream in the profession for at least a decade or two. There is a reason that visualisation is pre-rendered rather than realtime, and it isn’t necessarily that it looks better (well indirectly). It’s that it is tightly controlled and thus clients and or buyers can not find the bad bits and the fudged portions that will generally suck, also because drafting programs actually (even the BIM/3D ones) fudge a hell of a lot in the 3D sense to output 2D drawings which is still 99% of what buildings are built from.

    Real time visualisation will not take off quickly due to market forces not being able to actually be that ‘honest’ with enhanced perspectives making spaces seem larger and the like etc opening the way for litigation and the like when things don’t quite add up when it is built. An uncanny valley of sorts?? People understand a render is just that… a representation… as soon as you cross the line to virtually showing them things in 3D with freedom their expectations will be that much higher and thus that much harder to actually hit, let alone the exponential extra work required to get a model up to this level of scrutiny.
    The business already has next to no margins for production of visualisation so I can’t see this being profitable in the slightest except in extremely obscure instances and perhaps the experimental end of things.