Experienced Points: Is Google Stadia Doomed?

By Shamus Posted Wednesday Apr 17, 2019

Filed under: Column 159 comments

In my column this week, I talk about Google’s plan for Stadia, their new streaming games platform. This was a hard one to write because on the surface it looks like Google is being stupid or foolish. I’m comfortable second-guessing EA and Activision because they’ve perpetrated a lot of stupidity and foolishness in the past. But Google doesn’t have a history of this sort of behavior, which makes me think I must be missing something.

My favorite part is this bit from the Stadia website: 

I know I just said that Google isn’t dumb, but I have to admit this doesn’t rank very high on the Smarts-O-Meter.


Shamus Young and Robert Google are standing in a trade show booth beside a screen proclaiming “STADIA – THE FUTURE OF GAMING IS NOT A BOX”. Shamus is frowning at this screen. Bob is glad-handing people and inviting them to watch a video of a simulation of a Stadia proof-of-concept demo.

Shamus: What’s this about? The future is not a box?

Bob: (Grinning, animated.) Yes! Your games will no longer be limited by your gaming hardware. Instead, these digital dreams with come to you through the cloud, providing an innovative new experi-

Shamus: By “the cloud” do you mean, like, a data center?

Bob: (Somewhat deflated) Well, yeah. I guess. But these are very advanced machines!

Shamus: But no boxes?

Bob: Google’s Stadia technology can’t be limited to something as pedestrian as a simple box. This is pure gaming, delivered through a cutting edge network of fiber opt-

Shamus: Yeah, yeah. I get that. So in these data centers, you just have a bunch of circuit boards piled up in the middle of the room?

Bob: (Trails off, frustrated.) Well, no. Obviously. This is cutting-

Shamus: Cutting edge, right. But what do these machines look like? How are you storing them?

Bob: I guess they would be… on… racks.

Shamus: So you’ve got naked circuit boards on racks? Just all piled up?

Bob: Don’t be silly.

Shamus: So they’re organized somehow? Maybe you keep these parts in some sort of container?

Bob: I’m not at liberty to reveal the technological secrets we use to house our groundbreaking Stadia architecture.

Shamus: So what’s this container shaped like?

Bob: I can’t really…

Shamus: Is it a sphere?

Bob: (Frustrated, clenching his teeth.) No. It is not a “sphere”.

Shamus: Is it some other shape?

Bob: Maaaybe.

Shamus: How would you describe that shape?

Bob: It’s… boxy. A boxy kinda shape.

Shamus: (Pointing to the display screen.) So the future is a box!

Bob: Not just any box, though. With Google Stadia, the future is a box that is very far away!


I get that they’re trying to get us excited about the notion that we no longer need to buy expensive gaming machines, but even the Star Trek holodeck hasn’t transcended the need for a box. You still gotta put all of those fancy electronics somewhere.

The Fading Concept of Ownership

I didn’t like it when the world moved from disk-based games to content platforms like Steam. Sure, it’s more convenient. But I didn’t like the idea of paying for single-player games that depended on a third-party server giving me permission to run what I supposedly owned. I eventually accepted it because it was either that or find a new hobby, but cloud-based gaming threatens to erode the concept of ownership even further.

Shamus, have you not checked a EULA lately? You’re not actually buying software. See, you’re actually buying a license that just grants-

Yes, I know what a software license is, thanks. We’re dealing with two different concepts here:

  1. I own this ranch because I have a piece of paper that says I’m the owner.
  2. I own this ranch because I live here, I have the keys, and I’ll shoot any trespassers.

One type of ownership is concerned with morals, laws, and notions of property. The other is based on practical matters like control and physical access. The EULA is an attack on the first kind of ownership, and DRM is an attack on the second. This can get to be pretty complicated, since changes to one can be used as a justification for a change to the other. The DRM is excused because the user supposedly accepted the terms of the EULA, and users breaking the EULA is used as an excuse for why we need stronger DRM.

I don’t like seeing either of these two types of ownership attacked, but if I have to choose between the two I’ll take the second. I pay for my software, and I usually just roll my eyes at the various attempts at literal rules-lawyering in the EULA. “Fine. You idiots can claim whatever you want in the EULA, but I’ve paid for this software and I’m going to do as I please with it. No, I’m not going to buy a second copy for on my laptop. Piss off.” The EULA might state that I’m buying nothing more than temporary access to software I technically don’t own, but I’m still going to use it as I see fit.

But as our access and control is eroded, companies are gaining the power to enforce their ridiculous EULAs. Arguments over whether or not a EULA is legally binding are moot if the company can just press a button and make me stop owning something without needing to give me back my money. I don’t care what the EULA says, I’ll always think of that as robberyI realize this gets to be a little complicated when we’re talking about live service games. For the purposes of this discussion, let’s stick to single-player stuff. And yes, I realize even THAT line is getting blurry. Obfuscation and incrementalism are part of the strategy publishers are using..

I found the business model of the now-defunct OnLive to be really creepy. You were buying a game for full retail price, and yet you weren’t getting anything in return. No box. No disk. No software on your hard drive. I no longer have the keys to the ranch, and I no longer live on it. Instead I paid for the land and all I got in return was a promise from the actual owner that they’ll send me a picture of the place whenever I ask.

A Disposable Service for Disposable Games

I know these games sell well, but notice how they sort of vanish from the conversation when a new one comes out.
I know these games sell well, but notice how they sort of vanish from the conversation when a new one comes out.

Google’s Stadia might actually be better than OnLive. People are comparing it to Netflix. I’m not sure if this is an assumption or if Google is actually proposing a Netflix business model. I didn’t like the idea of buying a game on OnLive and also having to pay a monthly fee to access the game I boughtAgain, I realize this is how old MMOs worked, but I’m still talking about games that COULD run in isolation in the user’s machine.. But if we’re just paying the monthly fee and there’s no pretense of ownership, then I guess that’s better. Sort of.

It effectively turns the entire platform into a delivery system for the classic “weekend rental” type of games. It’s a game you launch, play through in a day or two, and never think about again. I think this is terrible for games as an art form and I’d much prefer to own a game, but I’ll be the first to admit that a lot of AAA stuff is just loud dross that isn’t worth owning.

Stadia is partnering with Ubisoft, and that seems particularly fitting. Ubisoft is working really hard to corner the market on disposable gamesEA’s annual sports titles notwithstanding.. I’ve played many Ubisoft titles over the years, but I only played them once. (And I often don’t finish them.) I think the last time Ubisoft made a game I wanted to own and revisit was Sands of Time in 2003.  That works out, since Ubisoft hasn’t wanted anyone to actually own one of their games in over a decade. Their idiotic devotion to excessive layers of DRM and Uplay hassles has meant that it feels like their games have one foot out the door at all times. Everything is just one closed server, locked account, or compatibility-breaking patch away from locking you out of the game forever. The premise of temporary ownership is built into their business model.

Assassin’s Creed? Far Cry? WATCH_DOGS? Tom Clancy’s Manshoots? Ubisoft’s AAA offerings are gaming’s fast food. Bland, predictable, and disposable. You don’t get attached to a particular McDonald’s hamburger, because you know that on any day of the week you can go and get another one just like it. If Stadia is going to be selling access to Ubisoft titles for a flat fee, then I can finally treat Ubisoft game like the rentals they are.

Google has promised to reveal more about Stadia – including pricing and launch titles – sometime this summer. I guess we’ll see then.

I should probably remind you to read the article, because none of this matters if Google can’t solve the all-important technology problems. These arguments about ownership and user control are irrelevant to most consumers, particularly younger folks who have no memories of the disk-based past. Those people are the future of the medium, and Stadia will live or die based on its ability to deliver them an acceptable gaming experience.




[1] I realize this gets to be a little complicated when we’re talking about live service games. For the purposes of this discussion, let’s stick to single-player stuff. And yes, I realize even THAT line is getting blurry. Obfuscation and incrementalism are part of the strategy publishers are using.

[2] Again, I realize this is how old MMOs worked, but I’m still talking about games that COULD run in isolation in the user’s machine.

[3] EA’s annual sports titles notwithstanding.

From The Archives:

159 thoughts on “Experienced Points: Is Google Stadia Doomed?

  1. Grimwear says:

    I don’t know where I stand on Google. They seem really smart but even then they have flops that get shutdown like Google+. Also is this something that people NEED? I mean didn’t that one Steam Box device do an in between of this where it would stream the game from your good computer to your ok computer in a different room? As far as I know it worked but it just wasn’t something that the masses needed and just fell by the wayside. More importantly if this is like Netflix then hurray…it will be amazing for a bit and then all the publishers/developers will hate the fact that they’re giving money to Google and will instead take all their games off and go create their own platforms.

    1. Tizzy says:

      A Netflix-style flat monthly fee would reinforce the notion of games as disposable. I’m curious to see if Google can even pull this off, as this seems to run counter to the publishers’ strategy of getting all of the money all the time. And right now, we’ve only just seen the first volleys being fired in the TV streaming wars. If Netflix-style game streaming takes off, we can look forward to long years of fragmented platforms ahead of us. The console wars will have nothing on this.

      1. guy says:

        I have to wonder how that would interact with microtransactions too. They seem pretty thoroughly here to stay, absent major regulation courtesy of EA’s screwups.

        1. Mephane says:

          I bet you’ll still have to log into your Uplay/Origin/etc account and the full range of MTX will be there for you to purchase. Like the header image so appropriately chosen, Assassin’s Creed Odyssey – a single player game for which they sell a goddamn XP booster, like they usually do in fremium mobile shovelware. Of course Ubisoft would want people to pay for that crap, too.

      2. Echo Tango says:

        You don’t have to look forward to this horrible, inevitable future. You could start a discussion group somewhere[0][1], or make a petition at change.org. Spread the word on your social networks, and take the (digital) signatures to lawmakers. As an example of positive change – cell-phones all[2] now charge with standard USB cables, and that was because a lot of people across Europe got sick of proprietary cables that end up in landfills, and all contacted their representatives in a large, organized movement.

        [0] Not here.
        [1] I’m told Reddit is toxic, but the platform itself seems fine. Small-scale discussions for Rimworld, for example, were conducted politely, and on-topic.
        [2] As far as I know, Apple pays a fine / fee, for using a non-standard cable.

        1. Sartharina says:

          Keep the government and politics out of this.

          1. Echo Tango says:

            Umm, yes, obviously? That’s why I included #0, and the link to the post explaining the no-politics rule, and linked people to other places so they can continue the discussion somewhere not on this website.

            1. Decius says:

              No, not “Get the government and politics into this by coordinating somewhere else”. Keep the government OUT of this.

              Yes, I’m near the ‘no politics’ line in saying this. But I’m on the same side of that line as you are.

      3. Blake says:

        “this seems to run counter to the publishers’ strategy of getting all of the money all the time”
        Not necessarily, if the publishers are getting most of their money from microtransactions anyway, then the cost of a lost sale doesn’t mean much if they’re still getting some subscription dollars and getting to sell DLC.

    2. Lee says:

      They seem really smart but even then they have flops that get shutdown like Google+.

      Worse, they have really great products like Inbox (or Picasa), which they then shut down.

    3. Chad Miller says:

      I’ve seen several accounts from Google engineers claiming that Google’s engineers’ incentives are not well set up to incentivize maintaining or improving existing services, and heavily incentivizes launching new ones. That would explain a pattern of launching a lot of projects which are then slowly abandoned.


      1. Joe Informatico says:

        My wife was ranting about this the other day. I’d say this is a clear sign that Google’s become a public utility, because democratically-elected politicians are always championing new initiatives, opening new rec centres and parks and libraries and such, but rarely want to deal with unsexy things like fixing decades-old infrastructure or transit systems, and this culture gets passed on to senior civil servants and bureaucrats and then on to staff. But frankly even corporations with a healthy market share who’ve delivered a quality reliable product for decades seem to be addicted to the “new thing”–or at least their shareholders are. (Or more likely, the rapacious private equity firms that are harvesting them for short-term profits and then bankrupting them.)

        Was there a clearer indication than when I keep getting prompts to download their new YouTube music app–when I’m already a paid subscriber of Google Music? And Google Music already includes YouTube videos in search results? What’s with the duplication of effort?

        1. Mike says:

          Actually, I have it on good authority that this particular screwup is just the result of Google not actually being a giant monolithic entity and actually, rather unsurprisingly, being a very siloed organization where people working on Youtube stuff don’t report to the same people as people people working on, say, Android stuff, until you get pretty darn high up in the org chart. Which means that the Youtube people don’t really know what the Android people are doing.

          Which is not to say they’re not incentivized to prioritize new things over improving old things, but it does explain the duplication of effort pretty easily.

    4. CoyoteSans says:

      “Also is this something that people NEED?”

      Yes, but the people who need it aren’t the people buying and playing the games like you or me. The people who need it are the developers and publishers who are desperate for any of these “games-as-streaming-service” to take off, as it will solve a lot of what they consider to be their most pressing customer service problems: finally uncrackable DRM that doesn’t interfere with the machine running the game, more controllable access fees, better metrics tracking, better ad targeting, and the ability to sunset more types of games more often to increase customer turnover.

      Notice how both this and Epic Games Store seem to be marketed more towards to the people making the games, rather than the people expected to play them? Or how despite the backlashes from fans, developers will not only lavishly praise these ideas, but go out of their way to defend them from the ungrateful masses? Developers people thought were the “good guys” suddenly seem awful willing to jump onto platforms that trample over consumer rights because it affords them a bigger payout or greater control over how their products are used and perceived (I believe developers see the lack of user reviews and forums on EGS as a highly desirable feature, not a bug).

      And if you, personally, think that’s horseshit, that’s fine: you’re considered ultimately expendable in the long-run. The publishers and developers are looking toward the bright shining future of the Fortnite Generation with Disposable Income: an entire huge market segment easily corralled through use of social shaming via self-expression and trivially bought “influencers” on social media. They believe these kids will one day become their ideal customers: willing to spend tons on microtransactions for insignificant cosmetics, willfully lapping up whatever their paid spokesmen are willing to feed them, and having no problems whatsoever never owning a single physical object in the game-to-consumer transaction loop. Before long, the old guard like us will be considered unprofitable old fogeys not worth the effort of even marketing to anymore.

      I believe 2019 will be remembered the year that game developers (not just publishers, but even indie and mid-size devs) showed their true colors and revealed just how much disdain and contempt they have for their customers.

      1. Echo Tango says:

        I don’t think the entire market would shift this way. Sure, lots of people don’t care about this stuff, and just want their cheap, stupid entertainment, but not everyone does. It’s easier than ever to create videogames[1], publish them, fund them[2], and spread the word[3]. If most of the games are trash or treat their customers like trash, indies or smaller studios can release products that people actually want to purchase, rather than being the only things available for purchase[4].

        [1] See Godot, Unity, the Unreal engine, the Doom engine, PyGame, Blender, etc.
        [2] Patreon, Kickstarter, Indiegogo, etc
        [3] This is ostensibly, the original purpose of all the various social networks – to allow people to freely share posts with each other.
        [4] Rental, games-as-service, whatever.

        1. Mistwraithe says:

          That is a promising way to look at it. The question is, are there enough people who do NOT just want their “cheap, stupid entertainment”?

          Look at free mobile games – I despise the Freemium model and won’t “buy” (for free) mobile games which work that way because I don’t want my kids being attacked by the gambling addiction model (and it is almost exactly the same – watch some GDC streams from companies doing Freemium games to see their thought processes on how to maximise revenue).

          I’ve heard the same sentiments from others on Twenty Sided. However, clearly we are such a tiny minority that it doesn’t matter. There are very few mobile games released these days which don’t use the Freemium model. I fear this will eventually be the same.

      2. Bubble181 says:

        I absolutely believe you hit the nail on the head.
        Stadia gives nearly-infinitely more control to developers and publishers, and youth today (oh God I feel old now) simply has a very different perception and notion of value of things like privacy, ownership, freedom of thought, etc.
        I’m not saying this as if it’s a *bad* thing per se – even if I don’t like it. But it’s just a fact that those aged 10-20 now live in a very, very different world from us.
        I was the generation that grew up exploring the Internet, this generation grows up shaped by a very different internet.

        1. Echo Tango says:

          The trading of privacy, ownership, and other freedoms without anything substantial to show/gain for it is definitely bad.

  2. Lino says:

    The only case where I could see Stadia succeeding is in the mobile market – although we’re starting to get mobile games approaching AAA standards, they’re mostly just pay-to-win dross. So I think there’s a space for good, AAA-quality games with non-predatory monetization business models that the Stadia could offer. After all, none of these AAA titles have mobile versions, and it’s easy to imagine publishers drooling for Google to essentially make mobile ports of their games (or to at least greatly assist them in doing so – after all, a major selling point is the ability to play on all your devices).
    However, this still leaves us with the latency issues – while AAA-style mobile games aren’t all that common, every single one I’ve played and seen has had amazing game-feel – something that’s impossible to achieve with high input lag.

    1. Higher Peanut says:

      What great mobile games have you played? I’ve always found them to be totally butchered by having to use touchscreen controls. I tried retroarch to run old games but could only stomach turn based ones. You can get controller shells for phones with real buttons but since they work wirelessly they chew battery.

      The one nice mobile game I’ve got is Cultist Simulator. It’s a port of the PC version that uses the touchscreen as mouse input.

      1. Lino says:

        I’m not saying the games are great – just that they have good game-feel – e.g. Brawl Stars (which I’m currently playing), there’s also Crusaders of Light (the game Diablo Immortal copied), and I’ve also seen some other competent-looking titles.
        Of course, they can’t compare to PC games when it comes to game-feel, but they still feel very fun, impactful, and most of all – responsive. The latter of which the Stadia will most likely struggle with.

      2. Marr says:

        Early days, but I have some love for Pixel Starships atm. It’s one of those things where you build a base for other players and alliances to raid, but it gets a million points* for said base being an FTL style spaceship from a Starbound side perspective. Success and failure is dictated primarily by player designed AI scripting for power management, weapons targeting and directing crew to man stations, repair damage and repel boarders so there’s a lot of depth there.

        * But then loses a thousand for describing itself as 8-bit. It’s 16-bit design, 8-bit was all monochrome sprites and double width pixels.

        1. Decius says:

          NES is 8-bit.

          1. Marc Forrester says:

            Exactly. The NES was the ultimate evolution of that architecture, managing to squeeze a whole two bits of color depth onto the screen. UFO 50 and PICO 8 are legit, but most games that claim 8-bit use an SNES art style.

      3. Gresman says:

        From the top of my head I would say Reigns, Plague Inc. and The Room are quite nice games, which got a solid PC port afterwards. Still waiting for the missing The Room games to be ported.

      4. Zekiel says:

        Device 6 is a fascinating, smart and atmospheric tablet game that could not work on a pc/console (because you have to rotate the screen to play it).

      5. Ciennas says:

        I never played KOTOR. Then it came to mobile, and I gave it a shot.

        Then I got it for XB1, and immediately regretted and switched back.

        That game controls much better with the touch interface.

    2. Lars says:

      Wait. ‘Isn’t Fortnite more popular on mobile, than on PC? I’d count Fortnite as a AAA title.

      1. Lino says:

        I know its mobile version is popular, but I don’t know if it’s more popular than the PC and console versions. After a quick Google search, I only found this – http://www.businessofapps.com/data/fortnite-statistics/

        Which basically just says that there are 45% more gamers playing on mobile than there were three years ago.

        1. Joe Informatico says:

          I remember one game journalist on Twitter writing about Fortnite, and she said it surprised her how many teens were playing who didn’t seem to care about trying to win, until she realized they were just using the game to hang out and chat. And when you consider how many children in North America (and possibly elsewhere?) aren’t generally allowed or aren’t able to regularly hang out with their peers in unsupervised settings (e.g. they’re bused for school and don’t live near many classmates or other people their age and don’t have a car), you realize that in some ways, Fortnite has replaced the playground, or coffee shop, or roller rink, or other public hangout for the under-17 set. And in that case, many of them probably don’t care how terrible the iOS or Android controls are.

    3. Echo Tango says:

      I’m pretty sure Stadia will be used to increase the amount and variety of predatory monetization practices…

    4. Mistwraithe says:

      If you think Stadia is going to have less predatory monetization then I fear you will be sadly disappointed. I expect it to be monthly subscription based, I expect the returns to the game developers for people playing their games to be low and consequently I expect most game developers to be heavily monetizing their games with in game purchases.

      It has been sufficiently successful (vastly more so than the traditional sales model – how do you think Fortnight, PubG, League of Legends etc make so much money?) already that I think nothing except possibly government legislation will stop exploitative in game monetization.

  3. Ander says:

    Stadia could be seen as a statement by Google. We have the processing power and infrastructure to handle the demands of gaming. Maybe they don’t have that, not really, not without Google Fiber. So fast forward a couple years. You know what they do have the power to remotely stream and process? Everything else. Why compile locally when Google’s servers can do it faster? Why store *anything* locally when Google has infinite storage? Of course I think there are good reasons to keep computing on your own device, but lots of people don’t care about them (they own Chromebooks after all). Maybe Google can’t stream AAA games with acceptable latency. But I bet they can stream most things people and businesses need a computer for. If all computers can be thin clients, Google would be delighted to hold the data centers doing the real processing.

    An aside on terminology: Shamus’s reaction to “no box” is how I feel about the term “serverless architecture.”

    1. Crokus Younghand says:

      For most people, the only computing device they own is a smartphone and cellular data latencies are not going anywhere. So, who would they be targeting? Enterprises? Those gnarly Excel macros don’t work on Google Sheets. Enthusiasts? They would abhor even the slightest of bad UX. Students? Those in first world countries are already using Chromebooks, those in third world don’t have the bandwidth. Artists? Any latency issues would make their life worse (and a lot of them are probably already using render farms, etc.). Developers? Remote debugging effing sucks, there’s a reason embedded development is so maligned (well, that’s one of the reasons).

      Who would be their target audience? Grandmas running XP? Who are they targetting?

      Probably the reason Google is going specifically for “Gamers©®” is because they might be the only one Google can milk, before the commoditisation of GPUs and better consumer grade hardware makes local gaming hasslefree (Switch might be a first-approximation of hassle free gaming). Maybe they realize it is now or never.

      1. Ander says:

        Those gnarly Ecel macros don’t need to work in Sheets. Stadia is Linux-based, and Google can host virtual machines of any OS you want. I’ve done a good deal of remote debugging in the last month, and while it does suck, that’s not what I’m talking about. The IDE (or whatever) could be on the server which could emulate whatever, probably better than we can locally. It’d be equivalent to remoting in to a computer to work with a compiler there, and you can do it from a cheap netbook or library workstation instead of an expensive personal computer without worrying about the server. That will probably always be at least as cheap as consumer gaming hardware; right now, it’s significantly cheaper. It will be less hassle for the majority of users. I suggest that “hassle free,” for a lot of people, would be logging in from any cheap but internet-connected workstation and accessing “their” fancy computer.
        Google’s audience would be anyone who can afford a Chromebook but wants to do work like they had a more expensive PC. That’s all Stadia is, but for a specific audience. I see no reason that audience needs to stay that specific. The limiting factor is still bandwidth, but the demands aren’t as stringent as they are for gaming. That’s why you start with gaming. Shoot for Mars, and you might at least end up on the moon.

        1. Cubic says:

          Your final paragraph … makes me remember the rock solid support commitment of Google. You send them an email (if you can find an address, that is) and nothing happens.

      2. Decius says:

        Google isn’t trying to get people to port their excel macros to Google Sheets.

        Google is trying to get people to rewrite the logic in those macros to work in Google Sheets.

    2. Retsam says:

      An aside on terminology: Shamus’s reaction to “no box” is how I feel about the term “serverless architecture.”

      I had the same thought, but I find that pedantic wrangling frustrating. Yes, there do exist servers in “serverless” architecture and there do exist boxes in the “no-box” future, but I think it’s clear-enough what both terms mean without needing to engage in endless pedantry on the exact names.

      The whole point (of both) is that, while the boxes/servers exist, it frees the user from having to manage them, which is a useful concept that’s relatively well expressed by the term “box-free” or “serverless”.

      1. Echo Tango says:

        The customer just has a small thing in their pocket, and all of the large headaches and loud computer-fans are somebody else’s problem, across the internet somewhere.

        1. Cubic says:

          Or it could be somewhere across the internet … but still your problem.

  4. Higher Peanut says:

    I don’t think I could possibly dislike streaming games any more than I do already. A total lack of control over games, even totally single player ones is terrible. No mods or tweaks and if a patch breaks a game or your saves you’re SOL. So much great content comes after release as people change how games play and add new content. Imagine having to play a Bethesda game with the default UI. Tying everything to your internet connection feels awful. Even if you’re in an area that can spring for the best connection that somehow never disconnects input delay in fast games can’t be removed.

    I’m also skeptical this is going to make games any cheaper. Going totally digital and massively expanding the market didn’t manage it. I think the best we’ll get is some subscription service/s (like I need more accounts for things) where you lose everything if you stop paying or step a toe out of line.

    The only use I could look forward to is unlikely to be utilised if it is even possible at that scale. Offload AI decisions in strategic games onto a better set-up than is possible for a home PC so that you can make AI play better and not just give them resource bonuses.

    1. Echo Tango says:

      Although mods *could* exist for games, I agree that they likely won’t – most games aren’t released with any kind of documentation or modding API. If devs now have to include some kind of Steam-Workshop-like thing, which allows users to upload mods, and share and use other peoples’ mods, it’s probably not going to happen.

      This is also going to be one more thing, which gets in the way of games preservation, for museums, etc. If Stadia is just a nice wrapper over something like Docker/containers, it would be easy to get it working in the future after the companies have all moved on or gone out of business. If it’s some proprietary server architecture, APIs and other nonsense, it’ll be an even larger portion of gaming history lost to the void.

      1. Decius says:

        Stadia-only games would be prohibitively difficult to mod in any case.

  5. Bloodsquirrel says:

    Google? Smart? Let’s not forget that Google is the one running YouTube.

    Out if all the companies in the world that I would not want to have the keys to the farm, it would be a creepy, authoritarian company with a track record of abysmal customer service like Google.

    1. Echo Tango says:

      Speaking of the people running YouTube – they don’t use enough bitrate to (reliably) handle videos of people playing games at 1080p. When it’s a nice boring game, sure it works, but if it’s a shooter or there’s any type of detail on screen, the video turns into a blurry, pixellated mess[1] instantly. So now they claim they’re going to do even higher resolution, for something that needs to be compressed on the fly? Hah!

      [1] Or Raviolis, as NorthernLion would put it.

    2. Erik says:

      I don’t disagree that Google has a bad track record, but it could be worse: it could be Facebook, which replaces “creepy, authoritarian” with “manipulative, deceitful, and actively malevolent”.

  6. Zaxares says:

    Unless Google can overcome the input lag issue, I think Stadia will ultimately be doomed to failure just as OnLive was. I think the biggest appeal for Stadia-hosted games will be MMO or PUBG-style multiplayer games that can host hundreds, perhaps thousands, of players simultaneously in vast open worlds, accessible wherever you are. But those sorts of games live or die based on lag, and if the player feels like the game is ‘sluggish’ or just not responding fast enough to what they want them to do, they will very quickly toss it aside. (Trust me, as a SEA gamer who for the last few decades has often been forced to play on US-based servers, either because it’s not economical to host SEA/OCX servers, or because I just wanted to hang out with friends, the gameplay experience can be borderline rage-inducing at times when you see other players do things that you know you could do just as well if the game would just RESPOND properly to you and on time. For a lot of younger players born into the broadband era who have never experienced that, I don’t think it’s a compromise they’d be willing to make.)

    1. Marr says:

      Persistent online worlds are definitely what this technology would be *for* in a sane economy. Trying to run such things partly client-side has been a disaster for many interesting projects, I was getting really into Worlds Adrift until they added a third party Windows rootkit system service to combat cheating. Stadia would bypass all of that.

    2. Ninety-Three says:

      Theoretically, input lag is manageable, because people already put up with a surprising amount of it. You can send pings across the Atlantic ocean faster than modern machines can put a pixel onscreen, and wireless controllers add even more lag. The best case for Stadia is that Google puts a datacenter in every major city: your wifi-enabled controller connects directly to a datacenter ten miles away (rather than sending data to your local box which relays it), it gets rendered instantly on some beefy server and bounced back to you faster than if you were playing on an Xbox. People have done the math, it checks out if Google invest in enough infrastructure.

      Datacenters in every city sounds wildly uneconomical, but Google have a moeny-printing machine in ads so I wouldn’t put it past them to pay for something that makes no financail sense. No matter how Stadia works, it seems to be that already.

      1. Cubic says:

        Google puts a datacenter in every major city: your wifi-enabled controller connects directly to a datacenter ten miles away (rather than sending data to your local box which relays it), it gets rendered instantly on some beefy server and bounced back to you faster than if you were playing on an Xbox. People have done the math, it checks out if Google invest in enough infrastructure.

        Datacenters in every city sounds wildly uneconomical, but Google have a moeny-printing machine in ads so I wouldn’t put it past them to pay for something that makes no financail sense. No matter how Stadia works, it seems to be that already.

        Corporations can blow money on ‘strategic’ projects for a while but doing something like that sounds crazy. Would be cheaper to send every subscriber a sealed box with that beefy server inside it.

      2. Water Rabbit says:

        I think the case of Google Fiber shows how unlikely that would be.

      3. Decius says:

        You shouldn’t be comparing the time to send a packet to the time to display a pixel.
        You should be adding *twice* the time to send a packet to the time to display a pixel. (one way for the input to reach the computer, one way for the video to reach the display).

  7. Psisigh says:

    I find myself less and less concerned with ‘ownership’ of games as I go on. I ‘own’ some two hundred-something games on steam, and a smattering on other platforms. However, if I lost all of my steam games somehow, I would find myself re-purchasing maybe a dozen or two for cents on the dollar relative to the price I originally paid. What’s the Witcher 3 cost on sale these days? I view so much of my gaming collection as a kind of disposable pile of stuff like a pile of trashy novels that are worth reading once then passing onto a garage sale.

    This is ignoring the fact that I could get months worth of gaming on EA’s subscription service, or similar stuff on Humble monthly.

    1. Marr says:

      The difference between Steam and Stadia is that with Steam the games do exist on our personal hardware, some of them are entirely DRM free by design and the locks on the others would quickly be opened by the Scene if Steam ever crashed and burned, it’d be a piracy renaissance. With Stadia you own literally nothing in the occupation and shotguns sense. It’s a whole other level of ephemeral.

      1. Dude Guyman says:

        To elaborate, the difference is that if a stadia exclusive game is taken off the servers and snatched out of your hands after you’ve already paid, there is no option to buy it again for cents on the dollar. Or get a pirate version, or do anything at all. It’s simply gone forever. Since nobody on earth owns it except google, nobody can stop it from ceasing to exist if google decides to delete it off their hard drives. Non-stadia exclusive games can at least be purchased elsewhere, but can still be snatched out of your fingers in the same way based solely on the whims of big brother google, whereas a steam game will only disappear from your local machine (or at least be unplayable without a crack) if steam itself goes under.

  8. Lino says:

    By the way, 2 years ago I stumbled upon thisvideo by Super Bunny hop where he presented Shadow – a service which is basically like Stadia, and that service is still around. The interesting thing here is that he didn’t experience any input lag. Give it a watch if you have the time.

  9. Thomas says:

    Google has built 7 separate messaging apps, and shut them all down again. They aren’t that sensible.

    They act almost exactly like a company with no strategic leadership. They build a new gmail, and then shut it down again a few years later. They have two teams building two copies of the same product all the time because no-one makes those decisions

    1. Crokus Younghand says:

      People in Google get promotions and bonuses for coming up with new ©exciting products and maintenance work goes unappreciated. Guess what kind of company culture that leads to.

    2. Echo Tango says:

      You mean like building YouTube Music, to compete (badly) against Google Play Music…instead of just improving their existing product? ^^;

      1. Thomas says:

        And then announcing to the public that they’re going to shut down Google Play Music and move everyone over to the (much worse) YouTube Music, and then not doing that but also not announcing that they’re not doing that anymore (or maybe they will).

        And don’t forget they officially merged the Play Music and Youtube Music teams in _2016_. So for three years, the same team has been building two different products to compete against…itself? And they’ve still failed to ship over half the features Play Music has.

        And lets not forget the pricing structure! For $9.99 you could buy Youtube Red which comes with access to Google Play Music, and also comes with offline access to YouTube Music. But! For the sweet sweet price of $9.99 you could instead buy a subscription to just Google Play Music, which only comes with access to one music playing service instead of two.

        They did fix that eventually. Now you can pay $9.99 a month for Google Play Music, or $9.99 a month for YouTube Music, or get Youtube Premium (formerly YouTube Red) for the brand new price of $11.99 which has both.


        Basically, Google heard the thing about infinite monkeys on infinite typewriters, and since they have infinite money from Google (Search), they decided to hire the monkeys to write their business plans for them.

    3. Lanthanide says:

      To be fair, they actually took a lot of the specific features that debuted in Inbox and integrated them into gmail directly.

      So the development effort was not wasted, but there’s no sense having 2 apps that (now) both do the same thing.

      1. Thomas says:

        The features they imported from Inbox aren’t implemented half as well as in Inbox, and there’s a lot of features they never ported over.

        When they released Inbox they told people it was going to stay too.

  10. Crokus Younghand says:

    Oh boy, wouldn’t Google executives be willing to sacrifice a newborn to live in an alternate universe where GPU tech doesn’t progress as fast as network tech does! When I can play Skyrim on a few year old integrated GPU, why would I want access to server farms?

    Regarding” Google-smarts”, you have to remember that these are in the end publically traded companies. With the next AI winter on the horizon (quick, how many consumer products use neural networks?), they need to put up new shinies for that sweet investor money. What Google product in the last decade has captured people’s imagination and than managed to satisfy the expectations?

    I am more excited because them trying to get in this business of paid services might be an indication that the ad/data economy that propped them up for so long might be drying out; but that might just be my pricacy-loving naivete.

  11. Geebs says:

    I actually wondered about the business of better multiplayer, while trying to come up with any positives of Stadia. Although the idea of having everything on the server is superficially appealing, it actually won’t work that way unless you exclusively play multiplayer games with people in the same postal district, because Google have tried to overcome the latency issue by moving the hardware physically closer to the player.

    This means it’s less like everybody logging directly into a single server and more like you’re just using a really, really long HDMI cable between your PC and your television. Also, your PC is locked in a cupboard, and you don’t have the key.

    1. Crokus Younghand says:

      Is there any data on where most of the bigshot youtube streamers are located geographically? Because if a majority of them are clustered together (which they might be, since they definititely have pretty good internet service), that might be the strategy Google is going for: have Stadia perform very well in those regions (e.g. give some sweet cash to ISPs), have the streamers all stream Stadia exclusive content on Youtube, then hope for social factors (fear of exclusion, etc.) to kick in. Youtube does wield a lot of influence over kids.

      1. Lino says:

        I don’t know – a lot of the gaming YouTubers I follow have started switching to Twitch, because the algorithm isn’t so kind to long-form Let’s Plays anymore. Granted, I’ve never followed the really big-ticket gaming channels, but even they have started making more streaming content.

        1. Crokus Younghand says:

          Which is why they might be rushing to launch this now ASAP. Better late than never.

  12. Matt says:

    Is there any hope that Kickstarter or similar crowdfunding could be used to preserve “games as a product” versus “games as a service?” It seems like studios increasingly use Kickstarter as a pre-sale or proof-of-market mechanism than an actual way to pay for their games being made.

    1. Lino says:

      I think there will always be a market for traditional games – we just need to stop hoping that EA, Ubisoft, and Acti-Blizz are going to be a part of it.

  13. Liessa says:

    I’m another one who prefers to own my games, and hates having to run them through a third-party client (to the point where I simply won’t buy platform-exclusive releases, even on Steam). I’d still be buying physical copies of PC games if shops actually sold them any longer. So I view Google Stadia as an interesting curiosity, but pretty much irrelevant to me. My only concern is that if it does succeed, it’ll gradually end up becoming the only way to play a lot of games (granted, mostly AAA games that hold no interest for me anyway) and it simply won’t be possible to buy them outright any longer. I’m used to waiting years on end for newer games to release on GOG, but a worst-case scenario is that this kind of thing could kill off DRM-free releases altogether.

    1. Echo Tango says:

      I suspect smaller studios and indies will continue to release without DRM. It’s a massive hassle for them and the customer, and simply offering a good product, at a reasonable price, easily (like a simple download on Itch.io) does a lot more to curb piracy than DRM.

      1. Decius says:

        One of the things driving people away from piracy is the reduction in credibility of .ru and dark web sources of executable files to provide them free from trojans.

  14. Hector says:

    The challenge I see here is financial, not technological. This doesn’t mean there are no technical challenges, but hypothetically with enough money you *could* make a reasonably solid experience (not necessarily an ouright good one). Most criticism I’ve seen elsewhere focuses on that side, while many appreciative customers think they’ll be getting every game they’ll ever want to play for maybe $15 a month.

    That is extremely unrealistic, and I can’t see that as sustainable even for Google. To do this, they need a lot of specialized hardware sitting around in data centers in ever corner of the United States (and then – THE WORLD!) They have a lot of data-crunching power, yes, but they can’t actually have every game ready for you to play at all times. They’re going to need to carefully curate games and will likely only be able to sell time on a selection of newer ones. I can come up with *some* ways around that, but they increase the time and money cost exponentially while diminishing the user experience. Worse yet for streaming experiences, gamers tend to be playing in a given area around the same time of day (evening) routinely, so there’s a lot less flexibility in terms of balancing your hardware across timezones.

    More to the point, no company will ever agree to this unless they are getting at least as much revenue per gamer as the alternative. If Ubisoft is signing up, then it’s because they’re getting paid somehow. Google might bankroll it for now, but in the long run they won’t cannibalize $60 full-price sales for a small slice of $15 a month. You’re going to have to pay up one way or another: either the fees will be quite high or you’ll still have to “buy” games, which will be available when and where Google says. Except a very disingenuous launch price.

    I’m not saying they can’t do this, but it’s certainly a very large hill to climb and I’m not at all sure that this is a viable product in the long-run, either because it won’t at the consumer end or it will bleed money for Google.

    1. Hector says:

      I know money isn’t the most popular of topics on gaming, so I apologize if brining up the economics annoys anyone.

      However, as per Shamus’s comment below about the hardware, I am extremely curious as to how Google intends to pay for all this. You need to have a box available to every player on your system in every market at peak time. And it needs to be able to play whatever game the customer wants, which gets really tricky unless Google pays a fortune for licenses, whether they site licenses or individual. And Google needs to buy, maintain, and upgrade/replace a LOT of hardware forever – with low resale value.

      Customers will be a lot less willing to pay if the service isn’t 100% available. And if they need to wait for data to be installed, its another hurdle. What’s more likely is that there will be some kind of rotating selection of games and only those titles, plus old games which are still popular. No idea how saves would work over time.

      I would be very interested in the kind of deals Google turned with the publishers. Is it flat-fee? Time spent gaming by customers? How will DLC work if the customers balk at paying separately and/or how does it handle customers with different packages? Moreover, Google engineers may be smart – but the answers could be different for every single game so how does that shake out?

  15. Ivan says:

    So, Im not sure how regular server farms work. Like, Google and other companies like data storage providers or whatnot have server farms. Those are made up of machines that consist of basically CPU’s and RAM and Hard Drives, yes? Basically. They don’t have monitors, or mice and keyboards, cos no one actually directly controls them, once they’re setup and running. They definitely do not have GPU’s, yes?

    I ask because a guy commenting on the Escapist article was talking about them (Google) theoretically using ISP’s local Servers somehow (he linked to a broken article to help explain what he was talking about too, which didn’t help me understand much) to distribute Stadia. I wanted to rebut by pointing out a regular server is literally incapable of running a modern game, since no GPU, but I realised I didn’t know enough to be sure my vague general understanding of how Servers work was correct or not.

    Someone help, so I can go back over there and reply like a smart person :)

    1. Shamus says:

      According to the GDC talk, Google’s Stadia machines are custom-built Linux boxes*. They’re partnering with AMD for the graphics hardware. I don’t think they gave specific specs.

      * A box? I was specifically promised no boxes!

      1. shoeboxjeddy says:

        These aren’t boxes, they’re… solid state cubes. It’s much more advanced.

    2. Retsam says:

      Tangentially, I don’t think it’s unusual for server farms to have GPUs already: a lot of server farms are being used to farm out work that’s highly parallel (cryptocurrency operations, AI training, etc), which are a good fit for running on GPUs, which is hardware designed for highly-parallel workloads.

    3. Decius says:

      It’s also plausible that Stadia would involve companion boxes being added to participating ISPs, with the GPUs that are being retired from bitcoin farming being added to those systems.

  16. GoStu says:

    My knee-jerk reaction is also to be worried about owning my games, and access to past titles… but I have to check that against the reality of what I’ve actually done in terms of using that stockpile of old games I own.

    When I made the transition from the original XBox to the XBox 360 I kept a whole stack of backwards-compatible games around… and I ended up playing the original Halo a little bit one time, and tried Morrowind but it didn’t actually work well. Now here in 2019 I just bought Morrowind on sale through GoG, so apparently I have no problem with paying again for something I already own(ed).

    My Steam library contains a few games I bought and never got into. Based on friend’s recommendations I got Hearts of Iron IV and found it incomprehensible; so that’s $40 I’m out forever. I snagged some Splinter Cell games on a sale price and never got around to playing them. Meanwhile there’s other games that I’ve thought might be interesting but weren’t $purchaseprice interesting, so I never tried them.

    My experience with other streaming services like Netflix has shown me that I’ll experiment with a show/game if it’s already included in a service I have; there’s been some gems that I would never have watched otherwise, but I tried because they weren’t any extra cost to me. Maybe if my Stadia subscription included both SHOOT GUY and PUNCH GUY I’d give those a shot.

    I’m in no way qualified to discuss whether it’s technically feasible.

    In the end I could see Stadia maybe working out for me, depending on the size of the library they get and what the price point is.

    1. RFS-81 says:

      But why even bother making an account if they don’t have GUN PERSON?

      1. Decius says:

        But be sure to boycott the service if they negotiate for an exclusive to GUN PUNCH MAN.

  17. DeadlyDark says:

    The next question to Bob Google – But what do boxes eat?

    1. WarlockOfOz says:

      Presuming that the technical issues are solved (big if, but not physically impossible) this could be attractive. While I’m not convinced it’s going to be better than earlier attempts (online, geforce now) the basic idea of timesharing a gaming machine rather than buying a whole one (and then only using it part of the time) has potential. Given Google’s record on starting then abandoning services I’m not going to risk any actual purchases but I might try if it’s pure subscription.

    2. Mr. Wolf says:

      Micro-chips, one byte at a time.

      I’ll see myself out.

    3. Decius says:

      Words, two bytes at a time.

  18. 0451fan0451 says:

    “The Cloud is just somebody else’s computer.”

  19. Mephane says:

    The Stadia hardware is distinct from all other gaming hardware, which means that Stadia is effectively a new platform and will further balkanize the gaming landscape with exclusives.

    Oh, now that’s a detail I was not aware of. I was expecting their server to just run either the regular PC, XB1 or PS4 version of the games, not a version ported to or made specifically for them. Not a fan of that prospect.

    1. RFS-81 says:

      Look like a huge gamble. They’re hoping that big publishers get their new releases ported to a new platform? If they could run older titles, they could prove themselves before asking for new games.

    2. Echo Tango says:

      Depending on how standardized and open the Stadia platform is, the games could just be running a slightly modified version of the Linux port. :fingers-crossed:

  20. Syal says:

    You don’t get attached to a particular McDonald’s hamburger, because you know that on any day of the week you can go and get another one just like it.

    You MONSTER!

  21. Xeorm says:

    I still stand by my first impressions on Google Stadia: it’s a waste to go after the hardcore gaming crowd by offering cloud based gaming with high specs. These are people that will go after wired input devices in order to reduce input lag. They’re not going to buy this service because it means an unavoidable drop in quality. I’m sure you’d get some people looking for budget gaming, but I’d agree that seems like a very small niche market.

    Instead, what they should have gone after was the budget market. Netflix of gaming isn’t a bad idea. Sell a large library as a subscription service with at least good quality hardware and you’d have a large market.

  22. Darren says:

    “Google doesn’t have a history of this sort of behavior”

    Google Glass?

    1. Shamus says:

      I didn’t think Google Glass was stupid. It’s not like it was a completely unworkable technology. It was just an idea that didn’t catch on. Same goes for Google+ and a few of their other canceled projects. Even YouTube and their search stuff are solid from an engineering standpoint. The problems people have with the platform is entirely in the realm of policy / company values.

      Put it this way:

      Glass was a fine piece of engineering, but it didn’t really find a market.

      Stadia seems like an unworkable idea from an engineering standpoint. That’s unusual for Google. (And maybe I’m wrong and they have a plan to deal with that 130ms of overhead. I don’t know. But based on what I know right now, I don’t see how this can work.)

      1. DeadlyDark says:

        You can’t beat the speed of light, yet

        1. WarlockOfOz says:

          Speed of light isn’t the issue, it’s the time currently needed for network handling. I don’t see any fundamental physical impossibilities in getting the ‘receive button press at server, update game state, render and return video output’ roundtrip down to similar delays as ‘receive button press at server, update game state, return for local rendering’. Plenty of technological and financial challenges, yes. Also and of course that roundtrip isn’t currently necessary in single player games.

        2. Echo Tango says:

          I actually worked this out this morning. Round-trip time around the world is something like 133 ms at the speed of light, and a typical distance between a bigger city and it’s smaller neighbors (i.e. the Google warehouses and all the people served, let’s say the width of Texas at 1200 km) would be closer to 4 ms. Pings for games in Asia are typically 500 ms or more, so the speed of light isn’t the limiting factor.

          1. Cubic says:

            I just did a traceroute to ‘www.google.com’, which turned out to be 14 hops and 116 ms. It’s not just photons flashing through a clear and private fiber.

            1. Droid says:

              Is traceroute a good metric, though? Pinging google is going to be a lot faster than using trace, as it pings every address along the way too, and adds that time to the respective pings, I’m pretty sure. At least I got less than 40 ms when pinging google and a lot more with traceroute.

              1. Cubic says:

                Well, we’re probably not in the same general location so I expect that’s why our times differ. If memory serves, traceroute just iteratively tries to send some sort of packet (ICMP?) with an increasing TTL. So in my case, it looks like the server was found in 14 hops.

              2. Decius says:

                40 ms ping times are about 10x the time taken by the light.

            2. Echo Tango says:

              That’s the point – even if you have a winding route through many cities and servers, the fiber isn’t your limiting factor. Processing the data, forwarding it, etc, can only be made so fast. Network switches are pretty fast, but they all add delay.

              1. Cubic says:

                Agreed, I was just pointing out there’s a lot going on on the way too.

      2. Darren says:

        Stadia doesn’t seem like an unworkable idea from an engineering standpoint in Silicon Valley. From my life experience in various parts of North Carolina and the internet infrastructure I’ve dealt with? They might as well be promising me a unicorn.

        I just fundamentally don’t see much difference between “tech that only Silicon Valley techbros want” and “tech that only Silicon Valley techbros can use.” They’re both a business failure because they have limited potential reach, and they seem like good ideas only because the people designing them can’t see beyond their own bubble to understand that it’s not feasible for a mass market product.

        It’s up to Google to prove me wrong on that one.

        1. Richard says:

          Perhaps, but given that it’s 30ms ping to nearest Google from London, UK?

          That’s 30fps-equivalent ‘latency’ assuming zero time for Google to render my actions and zero time for my PC to render a frame.

          So if we give Google really fast computers that only take half-a-frame (8ms), and assume that my PC can also decode and display the video in half-a-frame, that’s 46ms at absolute best.

          So 21fps equivalent at best, in a major capital city.

          I guess it could work for some games, but my brother just laughed and said No!

          Too slow for any FPS, too slow for KSP, too slow for Starcraft… KInda at a loss as to which games you’d bother with.
          Civilisation-style and turn-based games don’t tend to need high power graphics in the first place, so if my PC/TV/phone can decode VP8/H.265 fast enough to stream it, it can probably play it locally anyway…

  23. Paul Spooner says:

    Read the article on the Escapist and it feels like you hit the nail on the head. Loved the micro-drama here too. Well done!
    I’m kind of looking for the exact opposite of Stadia, an isolated experience with low-fi graphics that I can mod at will.

    1. Droid says:

      A wild Paradox Grand Strategy/Dwarf Fortress/Mount&Blade/Minecraft proselytiser appears.


      1. Echo Tango says:

        Also Rimworld. :P

        1. Droid says:

          Well, I meant “Dwarf Fortress” more as a genre rather than Toady’s pet project in particular.

          1. Decius says:

            “Dwarf Fortress” is a genre now?

  24. Agammamon says:

    But Google doesn’t have a history of this sort of behavior, which makes me think I must be missing something.

    Well, they sort of do. In that they often get a ‘good idea’ that is rolled out, becomes very popular, is very undersupported, and then is shut down with little notice because they can’t figure out how to make money from it.

    Stadia is going to be hardware intensive. There are going to be massive up-front costs here. The thing is, most of the world doesn’t have fast enough internet. Most of those who do are also the sort of people who pay a lot of money *deliberately* to have the best hardware (hence not needing a hosting service) and those that won’t – bought a console.

    So the only thing that Stadia really has to offer *in the short term* (some may choose to sign on over time in lieu of upgrading their hardware) is subscription game library.

    I can’t say whether it will or will not work in the long term – its not an inherently stupid idea (though I’m thinking bandwidth and latency isn’t where it needs to be in the majority of the world for this to work today)- but its got a big hill to climb up first.

  25. INH5 says:

    There’s another big technical problem: a lot of places in the US and presumably elsewhere have monthly data caps on standard internet packages. They’re usually large enough to not be a problem for typical video streaming, but they would definitely be a problem for this. I live in a major US city and unlimited data would cost me $50/month. $600 a year even if Stadia doesn’t have any monthly fees of its own. Or I could just buy a console for less than $300 and then make some of that money back by selling my now-redundant Blu-Ray player on Ebay.

  26. Bubble181 says:

    Man, there’s a lot of things I want to say about this, and not all of them are coherent with one another. I’ll try to stay somewhat focussed.

    1. As far as “boxes” go, I think you haven’t been in a Google datacenter in a logn time, and aren’t familiar with how they work these days. The days of “the cloud is someone else’s computer” is more or less over – the cloud is tiny little bits on a thousand other people’s computers. More practically, Google has a DC just a few miles from me; for a programming course I could look and see how some data flowed. You’d think most of the data would come from that DC to me and back, and updates were then rolled out over other copies in other DCs. That was how it worked 5-10 years ago – not anymore. You literally see one packet coming in from Mons (the nearby one), another from London, then one from New York, then London again. Smart provisioning and caching and whatnot mean this turns out faster than straight up streaming it from nearby. And not even in a P2P-network style, either – one software server is not at all limited to one physical machine (obviously) but even to one DC anymore. Playing on the “US” servers doesn’t mean you’re playing on either a server in New York or one in LA – it more likely means you’re playing on a server that is mechanically split over 50 cities. Anyway – the inside of a DC is as much “a series of boxes” as your house, or an office building, is, really. There’ll be racks there, yes…And your house is made of either bricks or planks, which are both just shapes of boxes, anyway.

    2. I absolutely hate hate hate the idea of (single player) games as a service, and Stadia is clearly the next evolution of that idea. Young gamers now have a very different idea of property, ownership, etc – and while our generation(s) might doubt and hold back, I can think of a LOT of younger people who’ll jump at the opportunity given a good price point. $25/month (comparable to Netflix), to play not only Today’s Hotness, but every month’s new Hotness? It’s practically free! And how many modern games still fall in the “worth coming back to 10 years later” category?

    3. In that vein – many parents would consider a Google-curated (more or less) marketplace with all the games the kid wants for $reasonable, a far better and easier choice than buying a beefy PC, and/or expensive console every few years. Heck, even non-parents might -the ever-increasing amount of people not just living on credit but living a leased life is proof enough. *Everything* can be a service. Philips is – I kid you not – selling lighting as a service now: a cheaper alternative to buying and designing lighting solutions for all your buildings, just have them do it and keep everything running.

    4. I play Diablo III on both the Eu and the US servers. Yes, there’s a lag difference, and those 60-100ms difference mean I stick to softcore in the US and play hardcore mode on EU only .Still, even the US servers rarely have over 200ms lag, usually somewhere around 120-140. The European servers, 20-40. That’s a game server, and Blizzard doesn’t actualyl have any hardware in Belgium as far as I’m aware. In the modern world – you know, not the outback of the US or rural Eastern Europe – lag is becoming less and less of an issue. Broadband everywhere, now with 5G starting to roll out it’s getting even better.

    5. Continuing from 4, lag is less and less of an issue…And, with many types of games, unimportant. True twitch-games will suffer, but I don’t think they’re the main idea. A LOT of AAA games is not that. Even outside of the obvious genres (I’m fairly sure I could still enjoy Civ VI with 2seconds of lag), plenty of games that are action-packed and reactive work just fine with slight input lag. Wireless controllers already cause almost as much lag. Wine or other emulators some more. Games adapt – many genres were supposedly impossible on a console without the fine control of a mouse, and it got “solved”. Aim assist sucks, but console players are used to it.

    6. I’m sure AAA games will in practice be a lot of AA games, too – games people are playing a lot. Is Minecraft a AAA game? No? Is it one of the most popular games ever? Can it run on a potato? Right. A $50 box that can play WOW, Minecraft, Fortnite,and a few big public drawing titles – The Division 3, Anthem 2, Horizon: One Dawn, what have you – will have people flock to the new system.

    7. I don’t understand how people here all seem to say there’s no market. As if everyone who has good internet magically also has the money, desire and capacity to also have a good gaming PC. If I draw a 200mile radius around my house, I’ve got Paris, London, Brussels, Cologne, Amsterdam, Dortmund, Bonn…there’s some 200 million people living here. Pretty much all of this area is completely saturated with high speed broadband and has 4G available everywhere. And yet, there’s plenty of people here who can’t manage to scrape together $1000 every few years to buy a good PC, don’t want to build something themselves, don’t have the interest or knowledge to keep it up and running…And who’d still like to play. Yes, that’s in large parts “console gamers” – but again, consoles tend to be expenive single-purchase items, which isn’t this era’s style. Besides, a €600 PS4 or €20/month, that’s 30 months before you’re breaking even…But two and a half years down the line, you might already be looking to replace or upgrade your console, or it mgiht have red ringed, or whatever.

    8. Google is often too smart for their own good. Technically, I still think Google+ was superior to Facebook. Doesn’t mean it didn’t fail. Google Glass was an idea ahead of its time – like VR in the ’80s or ’90s, it’s too early. But it’ll take off in a decade. Google tried to be in on the ground floor but was too early, just likethey tried to get in on the Facebook market but were too late. Sun/Oracle tried thin-clients-with-servers a decade or so ago and failed. It might fail with Google again, now – or it might be just in time to be in as the first big contender. If it does take off, thef irst one will have a HUGE benefit over the next one – just look at Amazon, Steam, Facebook, Youtube, WhatsApp, you name it. Eventually one or two big ones survive – definitely not always the first one (Friendster, MySpace,…Plenty of social networks before FB) but it’s a position of power *if* you can switch gears fast enoguh and follow the market.

    9. It’ll work, it’ll suck, and it’ll change both our hobby and the world.

    1. Echo Tango says:

      You’re assuming the computation can be split evenly between different data-centers. For multiplayer games, where the state of the game involves many players interacting with the world/monsters/objects, you need to do all the simulation in one place. Google could duplicate the simulation per player, so that the bandwidth costs are more local, but that multiplies the computational costs.

      1. newplan says:

        You don’t need to do all the simulation in one place – you can have plenty of stateless procedures that take inputs and return outputs but you do need the state data to reside somewhere. That somewhere then needs to communicate with all the data centers where the stateless work is going on and communicate with the player.

        1. Decius says:

          You can’t do the calculations asynchronously to reduce latency, because every player in a multiplayer environment needs to encounter the same game state.

          At least with current paradigms of gaming.

    2. INH5 says:

      I can think of a LOT of younger people who’ll jump at the opportunity given a good price point. $25/month (comparable to Netflix), to play not only Today’s Hotness, but every month’s new Hotness?

      You can already get any 2 game discs at a time through the mail for less than $25/month via Gamefly. Sure, swapping a disc out for a new one takes a few days, but does that really make much difference in the grand scheme of things? (And yes, you need a console too, but see my post above about the cost of that vs. raising internet data caps.) The fact that Gamefly never did for games what pre-streaming Netflix did for movies strongly suggests to me that “all you can eat” subscription models just aren’t viable for games for most players, probably because it simply takes too long to finish/get bored of a given game, whereas a movie is over in about 2 hours.

    3. Richard says:

      You would’t even be able to play Civ IV with 2 sec of Stadia-style lag.

      This isn’t ‘lag between where other player is and where I see them’.
      Stadia is lag between you moving your mouse and the pointer moving on screen.

      Playing Excel with a mouse lag above 100ms is quite annoying, and above 250ms is almost unusable.
      Word is perhaps more playable as you can use more keyboard shortcuts…

      It’s remote desktop on steroids – and has the same fundamental problem.

  27. Dreadjaws says:

    I don’t think a Netflix-style service for games can work. At least not in the current industry. Those problems you allude to in the article are even worse in other countries. I mean, sure, it can work as a niche service, but it certainly can’t have the kind of success Netflix is having. Netflix works because people have no problem with a movie or show’s visual quality suddenly lowering a bit when speed gets low, but for games things are different (and movies certainly don’t require continuous input).

    Also, if this service takes off and has any degree of success, prepare for the imitators following. EA is sure going to love a piece of that pie, and it won’t take long until some company decides to release their games exclusively that way.

    All of this is moot speculation because, again, I don’t think it’s going to happen (not anytime soon anyway). And let’s note that despite the popularity of services like Netflix home video sales are still high. People talk about the physical format as being dead, but that’s only really true for PC gaming. Movies, music and console games in physical format are still very much a popular thing.

  28. Dev Null says:

    I get that they’re trying to get us excited about the notion that we no longer need to buy expensive gaming machines, but even the Star Trek holodeck hasn’t transcended the need for a box. You still gotta put all of those fancy electronics somewhere

    I think you vastly over-estimate the average executive’s grasp of what cloud computing is. I have actually heard management say things like “It’ll be in the cloud *looks dreamily upwards while handwaving vaguely*. We’ll never have to worry about hardware failure again!”

    Right. _You’ll_ never have to worry about hardware failure, because when it fails, you’ll be able to blame someone else. But we also won’t be able to do anything to fix it, so the actual customer will be out a functioning service for considerably longer. (To be clear, I have nothing against pushing redundant failover servers offsite – it’s a core part of a reasonably robust design that no one ever wants to pay for. I object to the idea that one server far away is better than one server up close…)

    1. Echo Tango says:

      But what about ten servers far away?

    2. Decius says:

      One server far away *that is being maintained by someone who is in the top 1% of ‘maintaining servers’* is better than one server right here that is being maintained by someone you can afford to hire.

  29. Lanthanide says:

    I haven’t read any of the replies above so this might have been mentioned already, but this is definitely the way of the future, and if Google Stadia fails it will be for a reason you didn’t consider.

    Firstly, OnLive had a weird sales approach because they did not have the clout to negotiate with publishers for any other method of selling their content. The only option available to them was box sales, so that’s what they did. It’s like VidAngel with their censoring service for DVDs – you had to pay for a physical copy of each movie you wanted to watch even though they gave you access to the censored version via their website, because that was the only basis by which VidAngel could ‘legally’ manipulate the content in the way they wanted – there’s still a court case over this.

    Obviously Google has the clout to negotiate a different sales model with publishers, which is what they’ve done.

    Secondly, this is definitely the way of the future, for console gaming at the very least – as usual the PC master race will be a tougher nut to crack.

    Apple has already announced their own service, and both Amazon and Microsoft are working on their services and planning to release later this year. Sony and Nintendo appear to be out in the cold, at least for now.

    When all of the big names latch on to a new market like this, you know that it’s not going away. Valve caught everyone napping with Steam but they won’t make that mistake again. OnLive tried to replicate what Valve did but obviously lag-free streaming of games is much more of a technical challenge than simple digital distribution so they weren’t able to steal a march on the industry like they wanted.

    The most likely reason Google Stadia would fail is because they don’t have a back catalogue of games or historic relationships with publishers. Apple and Amazon don’t either.

    But Microsoft does.

  30. Drathnoxis says:

    Obviously, Google will use is repository of data they’ve been collecting on you for decades to predict your actions in game before you do them, thus circumventing any possible latency.

  31. Mr. Wolf says:

    Bob: I’m not at liberty to reveal the technological secrets we use to house our groundbreaking Stadia architecture.

    So it’s some sort of… black box?

  32. Blake says:

    I don’t think Stadia will work well for big AAA games played on a TV. I think having a console/computer locally is just going to be a much better experience for that.

    For slower paced games though be they adventure or turn based strategy, it might be totally fine if it saves you trying to download the game, find space on you HDD, etc. And for those sorts of games they should be able to stream to phones just fine too, which would mean you could start playing on your tv, pause, unpause on your phone on the train or toilet or wherever, and really feel like you can take the game with you.
    I think that would be the major advantage to a cloud based solution.

    If it went the subscription service route, it could also make browsing games much easier. There’s been lots of smaller games I’ve been curious about but haven’t wanted to have to buy it, download it, wait to install it, etc. If I could just click on it and have it load immediately that might get me playing lots more things.

    Having said all that I still think input lag will be a problem for most real-time games, an old housemate had a tv with bad input latency and I seriously struggled to play games on it. Playing over the internet would be far worse than that.

  33. I wonder if the latency thing is solved by intentionally slowing the other players down so that the latency of the slowest player becomes the latency for all (lowest common denominator).

    However there is still an issue, as a commenter on the Escapist article said, if you put the server next to your PC and ran a line around he world the latency would be around 138ms. Speed of light (or electrons) itself is a limiting factor, you are fighting reality and physics.

    This probably means that most gameplay will occur with regional (nation/county) or even local (same city) servers. Coop and multiplayer squad v squad might work okay (with the lowest common denominator latency as I mentioned) as data center to data center latency (peer-server-server-peer) is probably better (than peer-server-peer) as server to server traffic probably has higher priority on networks (aka “fat pipes”) and you won’t be accidentally routed through the neighbouring country for some weird reason.

    Now those 138ms are just “internet” (actually it’s more than that as routers and switches and bridges and such add some latency too), then there is the game client/PC latency (mouse click) and server processing (game engine) then the encoding and streaming back (and latency of the display itself which is 1 to 50ms depeding on how crappy a display you have).

    So playing with someone (coop?) halfway around the world almost guarantees a a 500ms (half a second) latency if unlucky.

    If both players end up connecting to the same game engine server and Google cleverly send world rendering data and render and encode at edge servers (you’d need special game engine and re-authoring of games to do this though) then that latency could be reduced a lot.
    Basically you’d have a game engine server that syncs stuff then edge servers has a copy of the game world and a small about of sync data is sent from the game server to the edge server which does the rendering. Since these edge servers are Google’s there is no need for anti-cheat software/middle-ware either.

    (Note: When I say “one edge server” I mean anything from “1 physical” to “1 cluster of servers”, with cloud computing 1 instance of a server is rather fluid).

  34. With more specialised games the edge server could render a large mapview of say a RTS game and simply crop the map to the view of each player and send that as the mapview data (with the player UI for that player overlaid), this would avoid multiple edge servers for multiple players needing only one edge instance and render.

    These won’t be normal games though and will only be available via Google’s Stadia, and probably their “secret” long-term goal.

  35. Redrock says:

    Shamus, why do you have to be so mean to Assassin’s Creed? Sure, the gameplay loop kinda justifies the label of “gaming fast food”, but the gameworld design deserves some credit. Or maybe a lot of credit. The Notre Dame story is a good example. Hell, my own trip to Italy was frequently punctuated by me going “huh, I climbed that” at seeing most landmarks. The AC Origins Discovery Tour was pretty good. What I’m saying is, I don’t think it’s entirely correct to lump AC together with even Far Cry, despite the similarity in the core gameplay loops. The historical settings and the effort Ubi puts into recreating them gives AC value beyond that of many sandbox games.

    As for Stadia, I’ll believe it when I see it, and that’s pretty much all I have to say about that.

    1. Shamus says:

      I kinda pick on AC because it’s Ubi’s flagship game and thus a perfect representation of all their faults. But you’re right. AC has incredible art. They don’t just have an incredible city like Grand Theft Auto, but every single entry features NEW (to the game, obviously) historical buildings in various parts of the world and at various points in history, all realized with incredible attention to detail. They even manage to turn these landmarks into jungle gyms without compromising their appearance. Amazing. Furthermore, the animation work is some of the best that’s ever existed in gaming. The team does that free-form parkour so easily and the result looks so natural that it’s easy to overlook just how difficult that is. (Has anyone else ever come close? I can’t think of anything off the top of my head.)

      I think the things that turn it into gaming junk food are:

      1) The near complete lack of compelling or memorable characters.
      2) An almost pathological resistance to doing anything interesting or clever with the narrative.
      3) Nothing really coherent in the way of themes or messages.
      4) A foundation of combat mechanics that aren’t really strong enough to carry the entire experience.
      5) Tons of under-developed side activities and novelty mechanics that feel like busywork.
      6) The typical AAA cutscene problem where cutscenes look fantastic but they’re poorly paced, the dialog is inefficient, and the scene is copying tropes without really knowing how to use them properly.

      I should add: I haven’t played any AC since Black Flag. I know that was a long time ago, so maybe the series has changed. My list above is how I viewed the series from AC2 to Black Flag.

      One of the points I’ve been dancing around for the last few years is that “narrative is more important than the publishers think it is”. (In this context, narrative = story, characters, dialog, etc.) Last of Us was a totally bog-standard shooter and people went crazy for it because it had a couple of really strong characters. Assassin’s Creed has all these things going for it, but the lack of an emotional connection or a central message makes it forgettable.

      1. Lars says:

        I’m playing Syndicate right now. The stealth system has improved a lot. The rest stays more or less the same. Your arguments still hold true.
        In the latest two entries Ubi changed much more. They changed the fighting system to something like Witcher 3. They added random number loot like Borderlands, including flame swords. They added damage numbers above enemy heads and character/enemy levels like WoW. They added Fantasy elements (a giant snake in Origin, Medusa plus Stone soldiers in Odyssey).
        I don’t want to play the last two entries. I can deal with this in a pure fantasy RPG, but it feels wrong for AssCreed.

        1. Decius says:

          It took me way too long to realize that you were playing AC: Syndicate rather than Syndicate (1993, Bullfrog) or Syndicate (2012, Starbreeze)

      2. Redrock says:

        Yeah, I can’t really argue with any of those points, nor am I inclined to. Like I said, I merely think that the way AC recreates historical periods and locations gives it more long-term cultural significance. I probably wouldn’t have brought it up if I hadn’t heard just the day before that some of the work they did to recreate Notre Dame for AC Unity may actually be helpful in rebuilding it after the fire. That matters, I think. I’ll actually take this further and say that, given the choice, I might very well prefer combat-free education-focused versions of AC to the usual stab-fest. Even though AC Origins Discovery Tour seems to be a one-off experiment Ubi ain’t willing to repeat with Odyssey.

        Far Cry, meanwhile, is utterly disposable.

      3. Blacky says:

        Easy parkour in Assassin’s Creed? You haven’t played in a long long time :-) Getting up/down in front of an obstacle with proper control still doesn’t feel half decent in 2019. Dying Light did it _much better_ in my opinion, years back, with a much much smaller budget.

        But yes, from the art and animation point of view, they do great work.

        1. Droid says:

          I still think the engine they had for AC 2, Brotherhood and I think Revelations was better at not annoying the player with nonsensical, counterintuitive movements than any of the ones afterwards. I remember being negatively surprised by AC 3’s mechanics, but thought it was mainly because of the inclusion of big cimbable rocks and trees, since that is where it happened most often. Seems they didn’t care or have the ability to improve anymore after that.

    2. Kdansky says:

      The Notre Dame story? You mean the part where a company with a revenue of 1.7 billion dollars per year spends $500k on a marketing stunt to garner some positive publicity? That’s the equivalent of one of us donating $15 or so. And they also give away digital copies of a five year-old game (which was badly reviewed) of their fast-food franchise?

      It’s a cynical marketing move. They don’t give a wet rat’s ass about the building or the history, the marketing department just saw a great opportunity to convert cash into advertisement at a great rate. Ubi spent $10 million on the Division 2’s launch marketing.

      And AssCreed is exactly that: Fastfood entertainment, designed to make you spend as much money as quickly as possible and then bore you as quickly as possible so you buy the next one. They even screwed with the leveling curve in the most recent games, so they can sell you an XP boost for an extra $10. That’s nothing sort of disgusting as a business practise.

      If anything, AC is Ubi’s least interesting series – the game mechanics are dross, the story and characters are nonsensical, and the whole thing is now bathed in RMT while doing the exact same thing for years. It has done nothing of value for gaming except produce some pretty screenshots. AC’s only positive side is that it looks good.

      1. Redrock says:

        I have to ask – did you donate 15$ to help rebuild Notre Dame? Or a single buck, even? I sure didn’t, so I don’t exactly feel like tacking up my high horse. Also, I’m not really talking about the money, money has little to do with AC. I’m talking about the fact that the work done to recreate Paris in AC Unity may actually be utilized to rebuild the Notre Dame. That’s some dedication to the craft right there. And it has nothing to do with whatever else corporate bullshit Ubi might or might not be doing.

  36. Rick says:

    No mention of PlayStation Now?

    I don’t know anyone with direct experience though.

  37. Noah says:

    Hey, Shamus! I’m a software engineer that used to work for OnLive, back when that was possible. I did analytics for them, so I’ve seen some of the financials for the things you’re talking about.

    We *did* try the NetFlix model for games, among others. Spoiler: it didn’t math out. One big problem that we had, that Google won’t, is that we had to make the business work by the numbers. They’ll have a much longer “feed money into the fire” runway than we did.

    But with that said, it’s not clear that consumers will go for *any* of the things claimed above.

    I can also tell you that while it’s technically possible to get a service like that to feel good, Google faces some harsh technical limits. Sony bought a number of OnLive’s patents (Google’s not allowed to implement certain important features that would violate them) but not OnLive’s trade secrets (which they could rediscover for a few engineer-years’ worth of salary, with the right engineers.)

    Since Sony *won’t* build a product technically equal to OnLive and nobody else is allowed to until the patents expire, you can pretty much prove mathematically that their service won’t degrade as gracefully as OnLive’s did, alas.

    1. Shamus says:

      Thanks for the insight!

    2. Lanthanide says:

      Patents can often be worked around.

    3. Decius says:

      Google also has the clout to show that the software patents are for things that aren’t patentable. Sony is likely to yield OnLive’s patents rather than risk losing a big chunk of software patents to the court.

      1. Lanthanide says:

        I seriously doubt that Google would try and bring down software patents for something like this.

        Software patents are used by the big boys as leverage against each other, as well as to prevent upstarts from competing with them.

        Far more likely that if Google really needed those patents, they would pay Sony money to license them (and Sony might ask for *a lot*), or they would sniff around and try and find their own patents that they can argue that Sony is breaking (whether they are or not) and use that as leverage to extract a reasonable licensing agreement from Sony.

        Or, as I suggested, just develop around them. Patents are for specific implementations, not concepts, so the same problem can sometimes be solved in a way that doesn’t violate the patent.

  38. Smejki says:

    Almost everybody is arguing whether Google can pull this off in technical terms. Response times, image quality, all that bullshit. I don’t care. My greatest fear -assuming the service will be subscription based – is that if the service takes hold all the games on offer will probably converge-mutate towards disposable endless grind-fest games-as-a-service. Why? Think for a second how Google will be splitting the revenues. It will probably be based on playtime. The more time you make people waste their time with your wannabe-games the more money you receive. A fantastic 10hour single player shooter enjoyed once by millions (think Doom)? Here are your pennies, looser. An endless provider of repeatable generated missions with basic game loop (think any looter shooter or any grindfest bullshit) enjoyed for years by hundreds of thousands? Here are you millions.
    Everything will be The Division/Destiny/AssCreed/Warframe/World of Tanks/Middle-earth: Shadow of X/Fallout76… you get the picture… because the others will not make enough money in this systems. All incentives work toward it.

    1. Decius says:

      Why would Google split the revenue to the sources that caused the most expenses?

      1. Smejki says:

        What would be the key in your opinion then? I really do wonder.

  39. Carlos García says:

    Heh, two weeks before the announcement I thought we’d need something like this, if only because most of the games I check or have need 20GB at minimum. That takes a lot of hard disk space and means I can’t have more than a handful games installed. While there are some games I can play once and then not be interested in replaying, I have several games I’d want to replay. And often I’m alternating playing among several. Sooner or later it would turn into a daily grind of uninstalls and reinstalls because yesterday I had the mood to play Blow Their Heads (75GB) and today I’m more in the mood for Do Menial Tasks Until I Can Beat The Demon (54GB), but later this evening I’ll want to play Conquer The World (80GB) and my hard disk has only 100GB for them.

    1. Redrock says:

      Additional disk storage isn’t exactly unavailable, is it? Even consoles come with a Tb of storage by default, and you can go crazy with a PC build. It’s annoying, sure, but not enough to justify moving to a streaming service. Unless, of course, you’re gaming on a laptop. But then, chances are, it’s either a gaming model with plenty of storage, or it’s not, and then it probably can’t handle games that require that much storage to begin with.

  40. Brian N. says:

    Remember the wisdom of Dick Masterson. Google’s a hardware company, they can’t innovate to save their lives. Every product they’ve ever made that wasn’t some kind of hardware was either ripped off from something else or bought outright and resold. Stadia is a piece of hardware, it does nothing to solve the intrinsic issues of streaming platforms for games.

    1. Lanthanide says:

      Strange claim to make, when Google’s largest revenue stream by far is from advertising, which has nothing to do with hardware, and Android, which is an operating system. They’ve only in the last few years started making hardware, including their own phones, and they’ve killed off quite a few of those products too.

  41. Blacky says:

    One thing wasn’t touched on, data mining. Off the top of my head, I can think of several juicy things that could be mined, like the (full) microphone audio tagged with the video and the input (especially in multiplayer games where you’re supposed to use your mic). How is IA/machine learning these days about interpreting grunts associated with known stimuli over a large, large userbase, I wonder?

  42. Fabrimuch says:

    I live in Latin America, where shipping fees and importation taxes means console and physical game prices usually get jacked up to double or even triple the price of what they cost everywhere else. That’s in addition to us making less money on average than other first world countries. As such, my perspective on digital games and Stadia specifically is that I would very much welcome it as it would make my hobby of choice way more affordable and I hold no real attachment to physical media as it is anyway.

    Buuuuut Stadia’s launch hasn’t been announced yet for my region. Figures

  43. Pinkhair says:

    Maybe Google has a different metric for success. Maybe this is about training AI off of user inputs, or about driving competitors out of business, or about pushing through infrastructure for other purposes.

Thanks for joining the discussion. Be nice, don't post angry, and enjoy yourself. This is supposed to be fun. Your email address will not be published. Required fields are marked*

You can enclose spoilers in <strike> tags like so:
<strike>Darth Vader is Luke's father!</strike>

You can make things italics like this:
Can you imagine having Darth Vader as your <i>father</i>?

You can make things bold like this:
I'm <b>very</b> glad Darth Vader isn't my father.

You can make links like this:
I'm reading about <a href="http://en.wikipedia.org/wiki/Darth_Vader">Darth Vader</a> on Wikipedia!

You can quote someone like this:
Darth Vader said <blockquote>Luke, I am your father.</blockquote>

Leave a Reply

Your email address will not be published. Required fields are marked *