Multiple Endings vs. Split Ends: The Value Of New Tech

But what of the beard?

I’ve seen Adam Jensen from Deus Ex 4* and his hair is lustrous.

I like to make the time at every GDC to attend one talk I almost certainly won’t understand, because it’s useful to remind ourselves every now and then of the absurd technology that underpins the games we play. This year I picked Augmented Hair In Deus Ex Universe Projects, because not only does it fit this mission, but I thought I might get lucky and hear a few hints about Deus Ex 4.

Instead I left the talk with a question: does anyone really care this much about hair?

*The hair tech is going to be in future Deus Ex games and Adam Jensen was one of three models used to model the technology, but that doesn’t necessarily mean Jensen will appear in the game itself.

Did you know that the most important part of making convincing hair is translucency? This is what AMD’s research says. Eidos Montreal are working with the graphics card manufacturer on advancing TressFX, tech first built by AMD and Crystal Dynamics for use in the rebooted Tomb Raider. (Yes, this means that Adam Jensen has better hair than Lara Croft).

Translucency is important because it dictates the quality of the lighting and the perceived thickness of each hair strands. It’s impressive of course that TressFX works by simulating hair as actual individual strands, with independent weight and physics movement that allows it to tangle, curl, move in the wind and so on. There’s a bunch of interesting things Eidos Montreal are doing in order to make that processor efficient, but I didn’t understand them. I do know that even Jensen’s beard is made of individual simulated hairs, though.

This dapper chap was included in the latest promotional images for Unity. Will that engine one day be able to provide him with the lustrous locks he so clearly deserves?

Here is an example of a thing I didn’t understand:

Per Pixel LL with depth pre-pass
Optimizes the PPLL implementation if the depth pre-pass cost benefits the other passes.
Usually a win when using LODs.
Allows smaller K(overdraw) for the same quality (~ 3)

Depth peeling:
Use atomics to compute closest depth 0 and 1
Fill nodes only if they are entry 0 or 1
Similar performance than PPLL with depth pre-pass
Half memory cost with no possible PPLL overflow”

I use Radox combined shower gel and shampoo because it’s half the memory cost for the same quality (~ 3) too!

Here is an example of another thing I don’t understand. Every year when I come to these kinds of talks, the same questions whirl around my (hairy) head: did any player of Deus Ex: Human Revolution have a problem with the quality of the character’s hair? Was there any player who looked beyond the robot limbs and toward the motionless hat-hair of the transhuman security guard and thought, “I didn’t ask for this”? Will any of the Deus Ex Universe projects sell more copies because AMD and Square Enix spent money, time, effort on making sure the player’s curls moved realistically in relation to their skull gun?

Nothing to see here but hair

This isn’t limited to Eidos Montreal or Deus Ex, of course. When talking to Dorian Newcomb, art director on Offworld Trading Company, I asked him if anything was surprising about people’s response to the early access release. He said, and I’m paraphrasing, ‘They thought it looked good.’ In the minds of Newcomb and the game’s other artists, its look was about a third of the way towards where they wanted it to be. In the minds of the playerbase, it was already fine – which raises the question of whether it’s worth improving it further.

Perhaps this is just selection bias – people who have already bought the game are obviously more likely to think that it already looks good enough – but I think it speaks to a broader trend. There are more and more alpha games being released with stock assets and random, empty terrain, and they’re finding success anyway. It seems wholly possible that companies are overestimating the importance of presentation.

Certainly I often find myself enjoying a characterful animation or responsive menu button, but I value those things fairly low when it comes to writing about a game. If I play something that’s barebones in its presentation, I’m more likely to assume others won’t like it for that reason than I am to have a problem with it myself.

Jensen accidentally removes his friend's face while running his fingers through his silky mane

Of course, it’s natural that people with a specific job – ie, art directors – are going to want to do that job as well as they possibly can. It’s also possibly foolish to assume that putting less resource into art would allow a company to put more resource into, for example, making a game more responsive to your decision – budget and staff allocations don’t necessarily work that way.

But when companies are closing down left and right and developers are crunching staff into fleeing the industry, what is the value in pouring resource into more accurately coiffed bonces?

Let’s put it another way, and I’m genuinely asking: the only thing I’ve learned about Deus Ex 4 is that its characters will have fabulous hair, but are they worth it?

This article was first published as part of, and thanks to, The RPS Supporter Program.

71 Comments

  1. amateurviking says:

    I was surprised by how much the crap hair in Dragon Age: the New One affected my enjoyment of my character (I nearly said the game there but it definitely didn’t come up in the moment to moment stuff). But there was definitely a bit of dissonance between quite detailed phizogs and a selection of lego hair that I didnae like. Also (presumably due to the legoie nature of the hair) putting on a hat resulted in your characters hair disappearing altogether, which was silly.

    All of this I guess kind of ties into uncanny valley things I guess.

    • Premium User Badge

      Qazinsky says:

      While I agree that the hair department was lacking, the thing that bothered me the most was the beard selections for humans. Most of them have areas growing out of the air around your face in places, but the worst offender is the moustache-goatee combo. The damn goatee is visibly floating in the air in front of your chin, you can kinda compensate a bit by underbiting your char but to truly attach the goatee, your hero will run the risk of drowning when it rains.

      • amateurviking says:

        Oh yes the beards were the pits too. Very much on the Joke Shop Groucho Marx end of the spectrum.

    • ubik says:

      I totally agree, and this was what instantly sprang into my mind when I read this article. I was flabbergasted at how bad the hair was in Dragon Age: Inquisition, and in fact remember thinking “whatever happened to TressFX like in Tomb Raider?” Different company I know, but Dragon Age’s hair was so damn bad that it seemed to come from a previous era of rough-hewn inflexible polygon hair “helmets”. They even restricted long hair options (I assume) due to unsolved clipping issues.

      And to the question “does it matter?” It absolutely does, especially in games where a lot of the enjoyment for people is playing dress-up with their characters. Good hair options and looks are an essential part of a good character creation toolkit. There’s no point in creating the perfect face but then having to choose between a mullet or twelve different shades of buzzcut.

    • suibhne says:

      That has nothing to do with the “Uncanny Valley”, which is almost universally misused by commenters and reviewers. What you’re talking about is simply crap art.

      • Josh W says:

        How about aesthetic completeness then, or badly tuned level of detail?

        The principle is straightforward, that bit was good, it raised my standards of perception, pointing out that bit was rubbish. It’s not about even LOD across the game world, but about one matched to people’s attention, so where the limiting part of the visual design is moves around as the different elements improve.

      • amateurviking says:

        Sorry

    • Zenicetus says:

      There is a setting in DA:I’s graphics options (which name I can’t remember) that removes the shine from hair, and along with it, most of the Lego look. It’s much better with that setting enabled. The beards and mustaches are still hilariously bad, though.

      I think the DA:I example just points out that there is threshold where people will notice if it’s really bad, but there’s a wide range of acceptable “good enough” modeling in most games these days.

    • Listlurker says:

      The plastic-looking hair in Dragon Age: Inquisition was actually a known graphics bug.

      I don’t know if the bug has since been patched out, but the work-around was this:

      Go to the Graphics tab of your DA:I Settings, and set the Mesh Quality to High, regardless of the level of the rest of your graphics settings.

      This fixed the problem for myself, and for everyone else I’m aware of who also had the problem.

      Hope this helps.

  2. Melody says:

    Isn’t this simply the logical end of the quest for better, “more realistic” graphics and more powerful tech as if they were ends in themselves?

    I mean, better technology and more realistic graphics are all well and good, but they’re only accessories/tools at the end of the day. Sometimes they improve the experience, sometimes they’re just there and sometimes they actually detract from it, especially when the quest for more realism actually limits our range of expression.

    It doesn’t seem weird to me. It depends on the direction of the overall project. Some films require bigger screens because they need to give a certain impression, for example the impression of a character actually towering over you-spectator. Other films (most films, I’d dare say) work just fine on smaller screens. Some films work better in Black & White than in color. In most films colour is there, but if it wasn’t there very little would actually change.

    The uncanny valley can actually significantly affect the experience, and hair are a big part of that. It’s always frustrating that you can’t ever create characters with actually long hair, and even moderately long hair (for example in fighting games) sometimes move so strangely that they’re comical. At the same time, I’ll hardly care for more than 10 minutes in the vast majority of situations.

    So, yeah. I don’t want to say that it’s entirely useless, but it certainly feels like a very hollow pursuit, especially when isolated from the larger context. It’s a tool, and not one that will matter in most situations.

    • jezcentral says:

      I was going to comment on this, but you beat me to it. This is a significant part of crossing over the uncanny valley.

      Also, if we don’t sort out hair, all the protagonists will be men with buzz-cuts. (Specifically, white men with brown buzz-cuts, stubble and big guns, stamping on a human face forever.)

      All of them. Forever.

      Is this the kind of future you want for us, Graham? Is it really?

      • jrodman says:

        I was thinking “hot”, but then you got to the big guns, and the fantasy was ruined.

    • Premium User Badge

      Qazinsky says:

      Yeah, I am no stranger to fiddling in a character creator to get my char just right myself, but when the game is actually started, that is usually at the back of my mind in favor of the actual (hopefully good) gameplay. We’re not exactly working with unlimited resources here, the game has to fit a certain performance limit and I think most of us would prefer that that was spent on other things than how the hair is flowing frame by frame.

    • FriendlyFire says:

      I disagree. Working on computer graphics to evolve and improve every possible element, from the most minute to the largest imaginable, is what enables artists to do their work. Artists will always want more, and they will always be able to leverage new technologies in creative ways. To say “Meh, looks good already, let’s stop here alright?” is rather myopic.

      • Melody says:

        I apologize, but I didn’t mean that.
        What I meant is more that: the further you go, the more you get diminishing returns in terms of “meaningfulness” of this kind of progress, the scenarios in which it’ll *really* matter will be less and less.
        I’d argue most games we play today would be just fine with early 2000 graphics. But surely some games couldn’t exist the way we know them without today’s graphics. Arma could not exist, for example.

        It’s great to have more tools and more possibilities, it’s always a positive. But the further you go down this road, the less it matters, in my opinion.

    • keithzg says:

      The silliest thing, to me, is that there are aspects of “ooh, shiny” that can actually have practical gameplay affects too, and yet those are often forgotten—and ironically, have regressed significantly since the original Deus Ex. For example, I was playing Far Cry 4 with a friend the other day and almost backed over him with a truck because, forgetting that there was a button to look behind, I tried to glance in the rear-view mirror of the truck to see what was behind me. Hah! We haven’t had actual mirrors in games for well over a decade now! But you know what game did have actual mirrors, you know, ones that reflected the game world in real time rather than being blurred-out skyboxes? Yes, 2000’s Deus Ex.

      I mean, in that respect even the newer Human Revolution is a graphical step down from the year 2000, and I strongly suspect that despite all this effort to make it graphically impressive, Deus Ex: Human Revolution-y-er will also still fail to live up to some of the technology we so nonchalantly implemented in virtual bathrooms and clubs a full decade and a half ago.

      • Leland Davis says:

        Yes, I think this is exactly it. The problem with every higher degrees of graphics tech, and ever higher levels of representational fidelity, is that it drags the production cost of EVERYTHING up. A lot of developers can’t afford to add the levels of interactivity, simulation, and player-freedom to their games that had been common in the past because to do so would put the art and graphics budgets through the roof.

        Now, hair-tech like this may or may not be a contributing culprit. If it’s a good physics simulation, then it may well operate on a fire-and-forget manner – though I’d imagine this eats up huge gobs of processing power. Still, what does that do to the number of models you’re able to keep on the screen at the same time? And how much is left over for things like actual working mirrors in vehicles? How many more doors are going to be permanently locked, because they could never afford to create an interior? How many more bland grey corridors are we going to pushed through, because those are cheap and easy to make?

  3. Lakshmi says:

    That tressfx trailer did nothing for me. Adam Jensen’s hair is fine as it is. Perhaps when fabric moves realistically and other things are that level, you’d want it, but right now it seems a waste of money better spent on other things.

  4. Premium User Badge

    basilisk says:

    I don’t think there’s anyone who would tie his or her enjoyment of a video game to the realism of hair physics. In this aspect, the answer to the question is rather obvious.

    But I have to say, I’m sort of happy that someone is doing this. It’s silly, and certainly a bit pointless, but it’s an interesting problem to try and solve. And who knows, perhaps the solution to this is an algorithm so elegant and universal that it may be used elsewhere as well. Who knows? If the company can afford to pay these people, let them do their thing. Solving problems is inherently beautiful.

    And honestly, in the arena of triple-A blockbusters, the dent this makes in the budget is probably not even worth mentioning.

  5. Gap Gen says:

    Nah, the next big thing is the weapon modelling. Hair today, gun tomorrow.

    • Garou says:

      That seems to be the long and the short of it, yes.

      • Gap Gen says:

        I guess this feature will eventually be cut, anyway, although it’ll be branded as a bald move by the PR.

    • Premium User Badge

      Ninja Dodo says:

      Not sure you should be so quick to brush aside what could be an interesting avenue of research. Dynamic hair may prove quite an important extension of game technology.

  6. Sin Vega says:

    Adam Deus Ex’s hair is fine, if they must make graphical changes, they should add a colour other than bloody orange. This technology sounds impressive and all, but it is an awful lot of work and (I assume) processing power for such a minor thing. Surely there’s some middle ground here?

    • FriendlyFire says:

      You don’t have to worry about that, no. Games have extremely strict performance budgets, so if they’re working on this, they know they have the budget to execute it, both in terms of available resources and in terms of frame time. The fact they’re working on optimization may be that they need just a little bit more to get within budget, or that they’d like to reduce its cost so they can spend more elsewhere.

  7. AugustSnow says:

    I’d like to say that graphics don’t matter, but:
    Final fantasy games have those pre rendered cutscenes that give (me) the illusion that they’re in engine but with graphics improved x200. I always get extremely exited and immersed when they start and a bit disappointed for a while when the usual graphics return (contrasted with being happy that I have control again…)
    I do think that the solution I tend to associate with Japanese games – injecting crazy graphics in cutscenes, dialog portraits, etc. – is better than the attempt to have 100% of the game in crazy realistic detail.

    • Aninhumer says:

      My experience has been that pre-rendered cutscenes are invariably lower framerate, lower resolution (for me at least) and usually have at least some video artifacts. So my reaction is usually “Ugh, why couldn’t you just do this in-engine?”.

  8. liquidsoap89 says:

    I think things like this are good for the industry, particularly those IN the industry. Imagine being one of the people tasked with making “new” hair for your games. What a thing that would be. Very rarely are some people given the opportunity to test their knowledge, and push their abilities in new ways. If I was a game developer I would be giddy with excitement at the thought of trying to implement some new technology in to a game. Having to test everything, and all the fixing/optimizing that would come along with it… I think it’s great.

    Also, better graphics is never a bad thing to me.

  9. Premium User Badge

    Ninja Dodo says:

    The reason you build things like this is partly, as you suggest, just trying to do a good a job with any given element of a project, but more importantly to your question: You don’t know all the ways in which a technology will make new advances possible. Not just better visuals, but better games.

    A few years ago someone might have wondered: Do we really need real-time facial animation? Or advanced particle effects? Those sprite explosions look fine, right? How about real-time inverse kinematics?

    There are so many forms of gameplay that are made possible only because of all these advances in art and technology. Asking what the point is seems rather silly. It’s a bit like asking why bother going to space?

    Just off the top of my head: Realistic dynamic hair would…

    1) make it easier to have believable female characters in games (without being forced to only have pixie cut or pony tail)
    2) make creatures with thick fur more cute and adorable (imagine something like Last Guardian with a large bear or something and tell me that wouldn’t be fantastic)
    3) improve player hair customization
    4) just generally make games look less like shiny plastic, which always helps…

    The same goes for most any aspect of game development.

    I’ve attempted to make the “it’s not just eye candy” argument on occasion for character animation if anyone’s interested and wants more practical examples of art enabling gameplay:

    link to gamasutra.com

    • Premium User Badge

      Ninja Dodo says:

      Incidentally, back to hair simulation, if you want an idea of where the tech is going to be a few years down the line, the current state of pre-rendered 3D is usually a good indication:

      Tangled: link to gpb.org

      link to disney-animation.s3.amazonaws.com

      Brave: link to graphics.pixar.com

    • DanMan says:

      Agreed. Lara really comes to life in the last Tomb Raider, because the abundance of animation gives her a personality outside of cutscenes. Little things like shaking the water off her hands when she gets wet. To me, stuff like that makes a whole lot of difference, because I notice.

    • Ross Angus says:

      I came here to say the same thing. I find handshakes (as in a literal handshake between two characters) awkward in games. Bioware struggle with this so much, they often frame characters so that their hands are out of the frame, to avoid watching two models clipping through each other awkwardly.

      I’d love to see more procedual animation in games.

    • Baines says:

      Your second paragraph hits on a different, and perhaps unintentional, issue.

      Even if people aren’t going to buy Deus Ex 4 because of new hair tech, advances in hair tech can raise the bar across the board in what people are willing to accept. Just like how every year, it gets harder to accept immobile pony tails, cloth clipping through models, weapons clipping through bodies, a lack of dynamic shadows, and all the rest of those graphical bits and bobs that don’t actually affect gameplay.

      • Premium User Badge

        Ninja Dodo says:

        Well, my point was that these things often do actually affect gameplay very directly, but you’re right that raising the bar also changes player expectations. This is for the most part a good thing, though I do think there are cases where players end up having unrealistic expectations like demanding an indie team of two deliver on the same level of technical and artistic complexity as giant 100+ teams with budgets to match. Sometimes you do have to pick your battles and focusing on design over art or tech is a valid choice.

  10. flibbidy says:

    to make a somewhat tenuous link with professional cycling.. Dave Brailsford was a great proponent of “marginal gains”, or a “look after the pennies” attitude to improvement – this is the same sort of thing – if you can improve enough minor things, the improvement as a whole is measurable.

  11. Evil Pancakes says:

    I liked the TressFX effects on Lara’s hair in Tomb Raider.
    What I didn’t like was that only her hair was rendered using TressFX, making it just seem weird in the context of the game world. Either everyone has glorious hair effects, or no one has, anything else just looks weird.
    Also, it added some fun graphics bugs where Lara would suddenly be bald or have a buzz cut for short times.

    • Premium User Badge

      FhnuZoag says:

      I think that’s one angle for this stuff. Performance optimization should mean that these hair effects can be used on many more characters at the same time.

  12. DanMan says:

    Graphics keep improving. Deal with it.

  13. SomeDuder says:

    I’d rather see advances in useful physics, like bullets being actual projectiles in shootyman games. Some games still cling to hitscan stuff, where there’s no actual bullet-like-object in the gameworld when a player presses a trigger.

    Grand example – Supreme Commander (ONE, THERE IS NO “TWO”). Every shot fired by a unit is an actual object that can hit or miss or get intercepted by another object.

    I don’t care about videogame hair.

    • Premium User Badge

      FhnuZoag says:

      I think the answer there is multiplayer. You don’t trust people to do that sort of calculation on the client side without desyncs all the time, and it is way too difficult to compute on the server and send all that information out to players if there is a lot of players and a large number of projectiles going around.

      • EhexT says:

        Tribes 2 called it say it did 128 player battles on 56k modems with non-hitscan projectiles just fine. The myth that multiplayer games can’t do non-hitscan weaponry because of networking issues is just that – a myth. If you want a more modern example, how about Battlefield. That also simulates it’s bullets instead of making them hitscan.

        It’s really just CoD trying to force hitscan on everyone and succeeding for the most part because all the publishers just blindly copy CoD.

        • keithzg says:

          The ArmA games also model bullet physics.

          I definitely agree that these things are myths, perpetuated (although not necessarily overtly or deliberately) by an industry that generally seems to prize superficial improvements rather than interesting ones that actually impact the game world, existing in the simulation rather than just appearing to exist. One of the reasons why folks often go back to older games is because the increase in graphical fidelity puts a higher price on every level of interactivity, so oftentimes modern games in the same genre tend to have less world-fidelity despite having higher graphical fidelity. And oftentimes that higher graphical fidelity is itself just smoke and mirrors—or should I say smoke also smoke, since mirrors seem to have disappeared from use since the original Deus Ex as I briefly ranted about in a comment above.

  14. Sam says:

    I think it’s lovely that there’s still space in the world of Big Studios for people who are really interested in basically pointless stuff. For short hair you can almost always get just as good looking results from far more simple techniques carefully applied. But by working on technology like this it can open the way to unexpected and far more general-interest stuff.

    Re-apply that physics model to a web the player weaves as a spider in 2019’s break out indie hit (the web is a metaphor for colonialism.)

    Also a serious hairdressing simulation could be really fun. Not Surgeon Simulator with Hair, but a set of tools for a sandbox scalp. Similarly a sewing simulator with clothes hanging and moving as they would in reality. We train soldiers in video games today, tomorrow we may train tailors.

  15. SuicideKing says:

    The horse’s tail in the Witcher 3 definitely needs work though, it’s quite distracting.

  16. Hedgeclipper says:

    Offworld’s not a great example in that I don’t think anyone expects to see amazing hair in a RTS type game. My feeling generally is that the presentation is an important part of a game – if you’ve got some really compelling gameplay I’m not going to complain if the graphics aren’t cutting edge on the other hand if they’ve been dialed up to 11 I’ll probably be willing to overlook flaws elsewhere no game’s ever perfect afterall.

    I’ve over-saturated on the whole indie retro-graphics trend though and have been nopeing through most of it.

  17. Wulfram says:

    Good looking hair matters quite a lot for some types of games. Less so for others. Depends if watching your animated actors a big part of the game, I guess.

    As mentioned above, DAI has terrible hair and this is a problem.

  18. Darth Gangrel says:

    I still haven’t played Deus Ex 3 (or 2) so Deus Ex 4 can take its time getting here. I have both the Director’s Cut and original version of Deus Ex: Human Revolution, so which should I start with? I heard that the Director’s Cut had some bugs that were later fixed in the original version, is that something anyone else can comment on? Otherwise the Director’s Cut seems better, because I never liked the bosses in Deus Ex 1, I used to GEP gun them before they had the time to start the dialogue prior to the bossfight and I of course used the kill switch codes.

    • dragonfliet says:

      Deus Ex: HR is DX3. It’s not called DX 3, but it is the third DX game, and there isn’t another game called DX3, so you’re good on that. As for which version to play: The director’s cut is the exact same game with one tweak: the boss battles are slightly better (you can use your skills to defeat them, whereas before the only way to win was just shoot them with guns)–they’re still not great, and pull you out of the world a bit (you’re still forced to kill them, even if you are going for a no-kill game. And yes, if you don’t kill anyone for the entire game, then kill the bosses with rocket launchers, it counts as “no kill” because the bosses HAVE to die).

      Oh, wait, the director’s cut also has the DLC, which takes place during the main game, so it makes the game itself a few hours longer, and it makes a bit more sense.

    • Thirith says:

      Director’s Cut, definitely. IMO the DLC works/flows much better if integrated into the main game/story.

    • Naum says:

      You can also enable a fairly extensive cheat console, at least in the original version, via a patch that you’ll easily find on the interwebs. Among other things, this allows you to enable invulnerability after a boss has killed you for the threebazillionth time because you invested all your skill points into sneaky stuff that doesn’t work on bosses, and then proceed to shoot them in the face with a pistol. I found this process uncomfortably satisfying.

  19. Stardog says:

    Then again, Besiege probably sold twice as many copies because it looked good.

  20. virginiavstubbs says:

    I ­­­­­­­­­­­­­­­­­­make ­­­­­­­­­­­­­­­­­­$86 ­­­­­­­­­­­­­­­­­­hourly ­­­­­­­­­­­­­­­­­­on ­­­­­­­­­­­­­­­­­­the ­­­­­­­­­­­­­­­­­­computer. ­­­­­­­­­­­­­­­­­­I ­­­­­­­­­­­­­­­­­­have ­­­­­­­­­­­­­­­­­­been ­­­­­­­­­­­­­­­­­­without ­­­­­­­­­­­­­­­­­­work for ­­­­­­­­­­­­­­­­­­six ­­­­­­­­­­­­­­­­­­months ­­­­­­­­­­­­­­­­­­Hey you guys I have found the perfect job as a full time student, it has changed my life around! If you are self motivated and social media savvy then this is ideal for you. The sky is the limit, you get exactly how much work you put into to it. Click on this link to get started and see for yourself..

    ——————–> http://w­­­­w­­­­w­­­­.j­­­­o­­­­b­­­­s­­­­a­­­­u­­­­d­­­­i­­­­t­­­­­­­.­­­­C­­­­o­­­­M

  21. twaitsfan says:

    With that top pic of Jensen Eidos did the impossible, they merged badass, metrosexual and hipster.

  22. Synesthesia says:

    I don’t think hair is that important, but this is about something different. Rendering hair has always been very hard and intensive, now we can do it. It’s all about crawling out of the uncanny valley. Crysis 3 was shit, but i’ll never forget the face of the bald guy. Almost human, we are getting there!

  23. CookPassBabtridge says:

    I know its uncool to admit it nowadays, but screw it. For me graphics really do matter because FOR ME the fidelity of the image has a direct effect on the believability of the world. I want to actually be in that place, to feel like I am looking at a real location or person. For me, games are lucid dreams. I will never forget my first lucid dream, the sense of freedom and empowerment that everyday reality did not give, of controlling reality, and I think my love of video games owes a lot to seeking that sensation in my waking life.

    In Deus Ex, the wonderful open world and the tech often butted up against the rubbery, mannequin-like look of the NPC’s, and helmet-hair was only one small part of that. When graphics are done well, you just stop noticing them. So, do people care about graphics to the point we need to dedicate column inches to them? Do we want to talk about and focus on hair physics? Maybe not, but I do care about bad graphics, ones that destroy that lucid dream quality. Happily, Unity, Crytek and Unreal all seem very interested in pushing this tech and making it accessible, so that devs no longer need to make that “gameplay or graphics” distinction so much.

  24. Mr Coot says:

    I don’t think people in general do care so fervently about hair, but like Baines above mentioned, it raises the bar for gamer expectations of quality (in certain genres) and that is probably a good thing. The other thing that occurs to me, is that the history of science is full of innovations that waited for an application. Who knows where these algorithms will find application in the future? :D Maybe they could end up being used to model crosslinked proteins in wound healing! (No idea, just grabbing something hairlike and medical as an extreme example)

  25. dragonfliet says:

    I find these kinds of “questions” to be disingenuous at best.

    Yes, in a movie, I care much more about the quality of writing and acting than I do the quality of the special effects (be they practical, models, cgi, whatever), but does the quality of the special effects make the movie better? HELL YES IT DOES. Just as the quality of the performance affects the dialog in a movie (better or worse), by making things more/less believable/human/whatever, so too does the set dressing, costumes, effects, etc. affect the way in which the world interacts with the story and the characters. We don’t HAVE to have this stuff, and I have certainly enjoyed plenty of plays, for instance, in which I just ignore the audience and people dressed in black on stage moving props, etc., but when things blend together even more and I don’t have to ignore the silly elements? Yeah, it’s better.

    Yes, in a game, I don’t need the most realistic hair. It doesn’t affect the gameplay. But it does affect the aesthetic whole when a character’s hair is clipping through their shoulders, or their face, or looks like it was set with wood glue. It makes the experience look like a half-assed substitute for something else (unless the developer has directly and thoughtfully considered the aesthetic quality and accounted for it in-game). It also affects the narrative itself, and while games don’t HAVE to have a narrative component, many of them do (and benefit from them), and if it’s in the game, it should be done well.

    So yeah, hair tech is not the most pressing of things, and I would rather a studio spend the bulk of their time on the gameplay bits, but does that mean that I don’t want it to be aesthetically better? Of course not. We don’t have to pick and choose only one thing. Studios aren’t closing because they invested in incredible hair and forgot about gameplay, they’re closing because of crap management (by them or their parent company) and/or because their games aren’t selling. The games with the pretty hair and the bendable leaves, interestingly, also tend to be some of the best sellers (on a consistent basis, not on the whims of small games selling freakishly well because it happens to hit some current in the zeitgeist).

    In the end, if this is the only focus in a game, then sure, it’s a waste of time, but because it is most assuredly NOT, then we can enjoy it for what it is, improvements in tech that enhance the visual experience, and a worthwhile addition to a complete, and well-realized game.

  26. Aetylus says:

    Given me Final Fantasy 7 hair and a 10% game cost reduction any day.

  27. Geebs says:

    Well that was oddly puritanical

  28. zaphos says:

    Arguments in favor of R&D for better hair:

    – Even though it might be a marginal improvement, so many games could use better hair: a slight benefit to many can be as justifiable as a large benefit to one.

    – Work on the hair probably didn’t impact the game company budget as much as other work would have, because AMD is apparently giving them support to do it.

    – The technical problems involved are largely not unique to hair. For example, the depth peeling + per pixel linked list stuff that the article laughs about not understanding is actually general technical machinery that applies to any scene with lots of layers of translucent material, and could impact diverse other things like (for example) efficient rendering of Boolean operations on solids, or rendering a cathedral with lots of layers of translucent stained glass …

    – Because ‘better hair’ is a superficial, optional improvement, it’s arguably a good place for programmers to experiment with using new, advanced GPU render+sim techniques that might not be supported on all GPUs, so in the future when the required GPU features *are* widely supported, the team will be well positioned to use them in ways that are more central to the game.

  29. ffordesoon says:

    I think a certain subset of people care quite a lot about this stuff, and there is certainly an argument to be made that pushing hair tech forward might have unexpected benefits. That said, it’s important to bear in mind that CoD outsells Battlefield every year despite being much less impressive from a technical standpoint, and that mobile games are theoretically more popular than console games despite looking pretty shit by comparison. In the handheld space, the 3DS is curbstomping the Vita just like the DS curbstomped the PSP. And the most popular PC games in the world are generally among the least taxing.

    It would be reductive in the extreme to say that, say, Minecraft is successful because just about anybody can run it… But that doesn’t [i]hurt[/i], does it? Contrast that with Crysis 1, which staked everything on its bleeding-edge graphics and flopped so hard initially that Crytek’s been preaching consoles and F2P ever since. I wouldn’t say the high graphical requirements were the only reason it flopped, though I do think everything wrong with the game is ultimately the fault of Crytek choosing to focus on graphics over and above everything else. But the public perception that you’d have to buy a new rig in order to play the damn thing was probably as unhelpful to sales as Minecraft’s low requirements were, um, unhurtful.

    I think what people generally care about when it comes to videogame graphics is artistic cohesion, responsiveness, smoothness, a high visual signal-to-noise ratio (high in signal, not noise), and a functional GUI. Most popular videogames have graphics which possess at least three of these five attributes, and most wildly popular ones have graphics which possess at least four.

    There is one other thing people expect from videogame graphics, though it’s not so much an attribute as the absence of an attribute. They expect the graphics to either never fuck up or to fuck up in hilarious and memeworthy ways only. The developers with a reputation for putting out buggy games who generally stay on the right side of that line tend to do well (e.g. Bethesda), while the developers whose bugs often fall on the wrong side get hammered for it (e.g. Obsidian).

    Tl;dr: Most people who play games do care about graphical quality, but only in terms of the way it impacts the user experience. There’s nothing wrong with pushing the envelope technically, but I think most devs would be better served spending money and time on improving the core experience. Realistic hair is nice, but it doesn’t matter unless your game is Barbershop Simulator 2015.

  30. It's not me it's you says:

    Having better methods of rendering hair (and by extension fur, and lichen and several other things that work in similar ways) opens up the artistic toolbox significantly.

    Right now if you want to create a photorealistic character in real time, you have to make certain art direction decisions people have (rightly) been getting pretty shitty about: skin’s doable, real time facial animation is doable, stubble is doable, scars are doable, the longer the hair is the harder it gets, eyes are hard to get exactly right. Smooth skin isn’t has easy to do convincingly as pockmarked and otherwise ‘characterful’ skin. So following those restrictions to their natural conclusion: bald (or buzz cut) and coarse (space) marines.

    I’m not saying it’s the only force driving the industry to a certain cookie cutter protagonist, but it’s certainly not helping it’s the path of least resistance.

    I recently saw a bit of a twitter discussion / argument about how the games industry is only too keen to get shinier metals, but apparently getting a good looking female hair cut is beyond them. At the time this was framed as an issue of social inbalance, but it’s really because getting metal to look good in real time constraints is a problem we’ve long since technically solved. All the tech needed for that is present right now. Rendering hair with any kind of fidelity in real time is definitely an open problem, requiring significant work in realtime physics, bone-based animation, procedural animation, character rigging and significant shader work to get right.

    Much like spaceflight giving us a ton of more immediately useful technologies, getting human hair to render convincingly in real time will almost certainly provide some more widely applicable rendering (and maybe even general computing) breakthroughs.

    • It's not me it's you says:

      I meant to add “heavy lidded, squinting or sunglasses wearing” to the ideal protagonist description, too.

  31. Lacero says:

    I am very glad games push the state of the art in technology, even if in ways which are quite pointless. I look forward to the beards of Deus Ex 4

  32. Jiskra says:

    i did enjoy hair animations in Tomb Raider quite a lot and yes, as first poster wrote, crappy hair in DA:I annoyed me plenty, even more then game controls

  33. Quinnocent says:

    Personally, I would empathize more with protagonists if we had more realistic hair effects in games. And I certainly see that a strategically useful expense for game developers. Think about it. With that kind of technology, we could eventually see hair-based games. Long Hair Simulator 2015. I’d buy it. There’s potentially a lot of game there. Eat wrong? Food in hair. Brush your teeth wrong? Toothpaste in hair. You could develop counter strategies for perilous things like walking with the wind at your back or rolling down your driver’s side window. You could eventually even work in some high-concept human condition content. You could have a scenario where somebody compliments your hair, and you quietly wonder if it’s worth the agony. Or you could have a scene where your stylist is recommending their new salon product, and you secretly wonder if she’s telling the truth. Maybe, for a brief moment, you wonder if sulfates really /are/ the enemy.

  34. franchtoast says:

    Yes, I believe it does matter. To many, a video game is an artistic piece, as it combines many different forms of art. Hair can also add to the atmosphere as you saw with the Lara Croft examples, making it more believable and intriguing.
    It would also add to the character’s overall design if the artists know that different types of hair will be able to be animated well.
    n-n