The Sunday Papers

Sundays are for childcare and comment moderation, sometimes more so than they’re for rounding up a weekworth of good videogames writing. Sometimes. This weekend we return to distract you from long weekend with some more good reads. Let’s be quick.

At Eurogamer, our own Alex Wiltshire writes about Mass Effect Andromeda and the quest for great facial animation. Interesting details from interviews with animators in this.

Cubic Motion, which provided motion capture for Horizon: Zero Dawn’s cinematics, produces algorithms called solvers, which take these scans and use them to interpret highly accurate and nuanced facial animation from motion capture data. “So you might take a smile, and break that down into ten different controls on the rig, and our tech then takes the actor’s performance and does a solve for each of those controls to create the animation.”

At the wonderful Waypoint, Jack de Quidt writes about Ghost Recon Wildlands, its beautiful world, and the suffocating effect of its ultra-violence, dumb politics and squad of hyper-masculine berks.

Right after beginning the game, I noticed a strange thing. Climbing into a car without waiting for my squad, I set off suddenly into the countryside and made it a short distance before, soundlessly, they appeared in the empty seats. I stopped, and got out of the car, and they piled out too. I climbed back into the vehicle and screamed away down the road without them, only for them to pop back. The squad will appear, faultlessly, in boats and helicopters and semi trucks. Sometimes they’ll begin an anecdote about a general who wanted to be paid by reference to how large his balls were. Sometimes they’ll open fire out of the windows.

Also at Waypoint, Jake Tucker on what it’s like to curate an indie games event, based on interviews with organizers of Feral Vector, Wild Rumpus and Now Play This.

Holly Gramazio, half of Matheson Marcault alongside fellow game dev Sophie Sampson, is the director of Now Play This, and has another audience to think of. Now Play This, part of the annual London Games Festival, has to appeal to casual fans and people drawn to an event billed as culturally significant and held at the capital’s Somerset House contemporary arts center. In other words, people who probably aren’t literate in the language of video games.

At GamesIndustry.biz, Nicholas Laborde writes that games have the power to change lives, based around an anecdote of a game he made that compelled players at its end to call a loved one. I am suspicious of thinking along these lines, as I think it often leads to well-meaning but shallow games which wear a mask of worthiness to mask minimal intrinsic value, buuuuuuuut it doesn’t need to be that way.

I’ve since become fascinated with the idea of games having a real-world component as part of their gameplay. This is not about a motion or fitness- based component (and is certainly not to discredit those games); rather, this is about causing real-world change as a result of playing a game. When the impact of Evangeline’s end-goal hit me, I can say that it was the closest thing to a “spiritual” moment I have ever experienced. I knew that I had to keep pursuing the idea of real-world action as part of game design. As Raconteur Games begins pre-production on our third game, we’re thinking of this real-world component and how our game can make the world a better place simply by players entering into our world.

Vanity Fair profiled Uwe Boll, infamous terrible film director of many game adaptations. He runs a restaurant in Vancouver now.

For someone known for his trolling and vituperative outbursts, Boll is also an instantly likeable and convivial host. We meet at Bauhaus, in the Gastown district of Vancouver, the restaurant he opened just before he decided to quit filmmaking. “That’s the wiener schnitzel,” Boll says over the pounding noises coming from the kitchen. “They have to hammer it.” There is one gigantic painting on the Bauhaus wall that says, in giant letters: “ART.”

Joel Goodwin attended Rezzed this year, and didn’t say hi. The bast. But he did write up lots of the games he played there, which you can find via this index post.

Radio 4 series Only Artists featured novelist and games writer Naomi Alderman in conversation with ceramicist Grayson Perry about computer games.

That’ll do. It’s a holiday, after all. Music this week is this live cover of a Nujabes track.

From this site

41 Comments

  1. Premium User Badge

    Ninja Dodo says:

    I feel like in the discussion about facial animation a lot of people (including many devs) are failing to understand that performance capture is not actually The Answer for games (or not the only one at least). It’s detailed and impressive, sure, but it is fundamentally non-interactive.

    “Ideally, no one would use generative tech for facial animation.” – Andromeda article

    Sorry, but no. If you want characters to react dynamically to what the player is doing you need dynamic procedural movement to adapt the actions and expressions of NPCs to what is happening around them. Basic things like looking at the player or points of interest, but also dynamic gestures and expressions based on real-time context. Capturing actors is great for cinematics, but for in-game movement you need something *interactive*. You CAN use capture for that, but it needs to be adapted to gameplay, either cut into pieces and blended together with transitions or adapted procedurally.

    The real solution is to *improve* generative animation and where possible blend it with more specific handcrafted or captured work. Currently the most interesting hybrid approach is Motion Matching (link to gameanim.com) but I’ve never seen it used for facial animation. Would be interesting though.

    • Unclepauly says:

      I don’t get how there’s still problems when Half Life 2 almost perfected it. Maybe not dynamically but for cut scenes I thought all later games would be at least equal to or better than. Yet we keep getting terrible lip sync.

      • Turkey says:

        I’m guessing it’s got to do with publishers wanting to own all of their tech to avoid paying royalties. So they kinda have to start from scratch instead of iterating on solutions that somebody else owns.

        • TillEulenspiegel says:

          The weird thing about AAA game development is that tech and art rarely get shared even in-house. This is far from EA’s or BioWare’s first crack at solving similar problems, and obviously they’ve done much better in the past.

          • Wulfram says:

            I think the problem was that they thought they could use facial capture technology to make the faces a whole bunch better and instead gave themselves a load of extra trouble.

        • Lacero says:

          In the eurogamer comments there’s the CEO of cubic motion popping up to say this. Of course he’s selling his stuff, but I think he’s right that hiring an outside specialist reduces the risk of something awful happening.

          • Premium User Badge

            Ninja Dodo says:

            I get where he’s coming from, given the article, but if you give in-house animators enough time to do their job properly relative to the scope of the task they are perfectly capable of producing good facial animation.

      • Premium User Badge

        Ninja Dodo says:

        Yeah, Half Life 2, and its Source contemporary Vampire Bloodlines still hold up surprisingly well as far as expressive characters go. Honestly, I blame the obsession with motion capture for dynamic procedural animation going largely ignored and unexplored since then, until recently.

        @Turkey: Yes, continual wheel-reinvention is definitely something that holds the games industry back.

      • Premium User Badge

        Grizzly says:

        I don’t think Half Life 2 and Mass Effect are really comparable there: Half Life 2 only ever has to take into account 10 or so faces, and most of it is just Alyx. Mass Effect has to make facial animations work for a virtually unlimited amount of faces due to the character creation. Despite that, Mass Effect 1 does a cracking job.

        (I have not played Andromeda yet so I wouldn’t know there).

        • Premium User Badge

          Ninja Dodo says:

          I suspect he was more talking about the general state of animation in games post-HL2, but yeah Mass Effect 1-3 did pretty well given their scope and complexity. The acting was never super detailed but it was effective enough.

          • Czrly says:

            I do not really agree. I think that the facial animation degraded dramatically between 1 and 3. 3 was just plain bad – with lips not lining up or closing properly and stuff. (Yeah, my first experience of the game was about six iterations of the character creator until I got exactly the right face parameters for the opening cinematic not to look like a pile of shit because of this.)

        • Ergates_Antius says:

          MEA may be doing more, but HL2 was 13 years ago. That’s twice as long as the gap between HL and HL2.

    • AngoraFish says:

      Funny how The Witcher 3 managed to pull off some of the best facial animation ever seen in gaming with absolutely no motion capture at all.

      • Premium User Badge

        Ninja Dodo says:

        They do use mocap, just not on the faces. They made extensive use of procedural animation to create the dialogue scenes, but they still used handkeyed and captured movement for individual parts and they would do unique complete animation for special more elaborate scenes.

        Basically, they had amazing tools, a large team of animators and a lot of time. link to gameanim.com

        It’s kind of a pointless exercise to directly compare them with Andromeda (without knowing their constraints and circumstances) other than to observe that Witcher 3 succeeded in a way that Andromeda did not and hopefully Bioware can do better in the future, either in updates or sequels (preferably both).

        Extra Credits’ Dan Floyd had a pretty good take on this as well: link to youtube.com

        • Kamestos says:

          Witcher 3 released 2 years ago, while MEA was developed. It stands to reason that the game in development would try to achieve a result at least on the level of quality of the former.
          Of course, budget and time constraints have an impact.
          But still, the result *is* disappointing.

          • Premium User Badge

            Ninja Dodo says:

            Well, that’s the thing: when Witcher 3 was released Andromeda was already well into development so it would have been way too late to go “let’s change our entire animation pipeline and do what those guys are doing”, not to mention it wasn’t until even later (GDC 2016) that more information became available about just how CDProjekt were able to achieve what they did. What’s more surprising is that they were not able to get more of a head-start from previous experience with the Frostbite engine on Dragon Age Inquisition (at least sharing information between teams in so far as there was no overlap) as that game did not seem to have those problems so much. Judging from the outside it sounds to me like they simply didn’t get enough time for the scope they were going for, but let’s hope they manage to address some of these issues in future updates.

        • Synesthesia says:

          Thanks for the link! I always wondered about The witcher’s animation work.

          • Premium User Badge

            heretic says:

            Ditto! Great link, nice explanation – subscribed to that channel, looking forward to other features like this, very interesting to get a closer look at how games are made.

    • Lacero says:

      Generative eye movements, idle anim blending, weight shifting etc. are absolutely needed and it really saddens me that mocap became the answer to all this.

      Even when you have mocap you still need to bend things properly and this is actually really really hard too. Too often people draw hard cut offs for the blending, and you end up with mouths lip synching while the rest of the face is acting something totally different.

      Another common one is people running with their legs while their upper body is shooting, no shoulder movement of weight distribution on the top half at all. A lot of cheap MMOs (and some expensive ones) do this.

      There’s a place for mocap, mouth shapes, specific complex movements, but to me it’ll always be a starting point for clever, well tested generative code/state machines. Or it’ll look bad and/or be too expensive to use everywhere.

    • syndrome says:

      Squadron 42: Facial Animation Technology

      as far as I know, made in Serbia

  2. Ghostwise says:

    The first (and for me, last) paragraph of Wiltshire’s article may have led to an utterance of something not fit for polite company.

    If that helps:
    – soldiers are real people who exist. They’re not video game or movie characters. Comparing bad feedback over an office job to having your buddy killed right next you is… is… I don’t have to explain it, presumably ?
    – said paragraph mentions people posting clips of defective animations, then moves on to denounce “indefensible attacks”. Either there’s an important bit missing, or a switch decaf should be considered.

  3. Bobtree says:

    I found “Why Sega is obsessed with Humble Bundle” a surprisingly good read: link to gamesindustry.biz

  4. Zenicetus says:

    The Wiltshire article mentions something I’ve often heard as partial defense of what happened with Andromeda — that it has more lines of dialog than ME2 and ME3 combined, and 1,200 speaking characters that are double the number in ME3.

    That’s a dodge, because it’s only the human characters that are so terrible. The aliens are carrying at LEAST half the dialog in the game, maybe more. And they’re fine, due to lack of Uncanny Valley and the fact that Bioware is using the same face models and animation for all the aliens. Except for the main team members, every Asari, Turian, Angaran, Salarian, and Krogan has exactly the same face in the game with different paint jobs. It’s especially noticeable with the Asari because they’re the closest to human norms. Maybe it’s rose-tinted glasses, but I remember the NPC/non-team aliens as having more variation in what they looked like, in the previous ME games.

    So the game was already heavily streamlined in that respect. All they had to do was get the human faces right, and they blew it.

  5. Chaoslord AJ says:

    Regarding Andromeda: either the animators blew it or qm, sales, executives.
    I’m not hurt, I’m not angry like “they killed my favorite franchise”, that are way too strong feelings for a piece of entertainment (an issue which could be discussed seperately).
    Just don’t expect me to buy this game full-price.
    BTW I perfectly understand their desire to make excuses, those gifs really ridiculed them.

  6. poliovaccine says:

    No offense meant, but if the ending of Evangeline is the closest thing you’ve ever had to a spiritual experience, well, someone needs to start you off with some psilocybin mushrooms, or at least a nice hike in the woods or something, hot damn…

  7. Dogshevik says:

    @Andromedagate

    If people keep buying rushed, unfinished products they will be given rushed, unfinished products. That is all there is to it.

    The rest is just the usual background noise.

    Shrug.

    • Zenicetus says:

      The rushed and half-finished release only accounts for a few problems though, like the lack of QC on animations and a few bugged quests. Although not many of those. Overall, the game is pretty solid, which explains why they released it when they did. On a purely mechanical level (and ignoring the terrible UI design), I’ve had almost no issues with how the game functions.

      The main flaws were baked in, long before the final rush to release. Like the laughably juvenile dialog writing, the indifferent voice acting, frustrating UI, and the repetitive combat padding. With much better writing, a better UI, and a game half the length, this could have been a good successor to the series.

  8. Sly-Lupin says:

    I definitely think there’s a lot to be said of media’s capacity to change lives, even though I really don’t understand it myself. I don’t really think of myself as a “game developer,” but I am a writer and I have experimented some with interactive narrative media (basically text adventures). I’ve had people contact me based on their experiences with my “game,” which evidently had a profound and resonant impact on them. I’ve had essays drive people to tears.

    But I look at the content I created that incited these reactions… and I don’t see it. I think people project a great deal of themselves into media, and probably even more so in interactive media. And that investment enables deeper emotional connections and greater transformative experiences–regardless of the creators’ intent.

    So, yeah: it’s a very interesting topic.

    • Premium User Badge

      Seyda Neen says:

      In my view, to consume art is to bring everything you are to the table in a conversation with whatever is being consumed, and that’s a beautiful thing. I don’t think it’s possible to not project yourself onto art, although trying to leave a space in your mind for an author’s intent, with less projection, is still a worthy goal. But I think some people try to minimize their projection as much as possible and that’s boring to me.

    • poliovaccine says:

      One thing I’ve learned for myself is that, when you create or produce something, a song, a game, a picture, whatever, you’re releasing it into the world as its own entity – your intentions for it are frankly incidental, it’s entirely there for people to project themselves into, and in a way, that’s where the real value of creative work comes from – i.e. not your own experience as the creator, but the range of experiences your creation offers to others, more than you might have ever conceived or intended. It’s there for them more than for you.

      I used to think it was a bit of an unfunny joke that I could write an ironic, self-deprecating song about desperate heroin addiction only to watch people miss the double entendres so deeply as take it for “such a sweet love song!” But these days I see that as only adding to the song’s value, if anything (with that other meaning remaining buried but available to anyone properly acquainted with the slang to hear it). If your creation is put together in such a way that six different people can get six starkly different, yet equally affecting experiences out of it, that just means you’ve successfully imbued it with that much more of its own life.

      But yeah, it used to drive me a little crazy when I’d see people enjoying something I’d made whilst simultaneously “missing the point.” In retrospect, of course that would drive me crazy, because there is no “point” to self-expression, it’s an end in itself, not a means. I’ve also had the experience of someone being affected emotionally by something I considered that I’d sort of lazily tossed off. Again, at the time they just seemed like a sentimental kook, but in retrospect I was just giving my own creations less than their due respect. Those people aren’t “wrong” or somehow mistaken for self-projecting… if anything, that’s how this whole “connection/shared experience” thing works.

      So yeah. I totally get that angle, but I wouldnt say that degree of self-projection is necessarily mistaken or out of place – I kind of think that’s what’s happening anytime two people communicate at all, on some level, though of course only moreso when the things they communicate about are nonphysical and/or abstract – as they tend to be with creative works, at least to a greater degree than instruction manuals.

  9. porcelain_gods says:

    Something I thought a great read about Jeri Ellsworth

    From hacker to Valve and back again
    Race cars, computers, hacker toys and augmented reality
    link to polygon.com

  10. Chromatose says:

    Jack De Quidt’s article about Ghost Recon Wildlands makes me hope somebody develops a mod that just deletes all of your squadmates and enemies and allows you to go free-roaming through a virtual Bolivia to your heart’s content. The game looks completely beautiful, but I absolutely do not want to play it.

    • The Chadillac says:

      Then don’t. Why don’t you consider going on an actual vacation to Bolivia instead of wasting your time looking at a shoddy computerized facsimile of nature? Do you really think you’ll find some kind of sublime beauty in a fucking Ubisoft game?

      • LearningToSmile says:

        Ah yes, because a vacation in Bolivia costs somewhere in the ballpark of $60, so it’s totally a reasonable choice compared to a videogame, right?

      • Chromatose says:

        Because I believe that virtual recreations of real places have genuine creative and artistic merit. The concept of virtual spaces is IMO one of the most important contributions videogaming has made to popular culture, and I’m not going to be dissuaded from that viewpoint by some dweeb in a comments section making needlessly angry retorts to points that needed none.

  11. LearningToSmile says:

    Wasn’t that Uwe Boll article already posted? I distinctly remember reading it before.

Comment on this story

XHTML: Allowed code: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>