Google Stadia

Google's keynote speech at GDC has revealed their entry to the games market, and it's not a console – it's a streaming service.

It obviously all looks very slick and cool on stage, but… there's no way it can work that well in real life – is there? If it was anybody other than Google I would undoubtedly be more skeptical, but they have the resources, engineers and –perhaps most importantly – the cash to throw at this thing, and I suspect they might just pull it off.

Or it'll get unceremoniously shitcanned in eight months and its third-best feature will get included in Gmail or something.

They've been working on this for a long time; the latency issue is, essentially, gone. If you live within a fifty miles of a major city, you live within 10 ms of a google node which is where this stuff'll get handled. I've not seen it in motion but I know that there's a lot of people betting on this.

Not for me, I'll pass. Not saying it won't be good or work, but it's not for me.

When it runs on my Apple TV I’ll have a look.

I’ll be waiting for a while then.

Not for me, I'll pass. Not saying it won't be good or work, but it's not for me.

Why not. If it works and it's seamless it's how we'll all consume games fairly shortly. We're already miles down the road away from the thrill of a disk product and manual sniffing.

It’ll be interesting to see how it goes. It’s definitely a different vision of the future to current systems, which are increasingly about amazing visuals and high frame rates. This will never have the same kind of visual fidelity as a proper home console; the question is, will people care?

I’ve used the Steam Link a fair bit for local streaming and even in a best-case scenario (dedicated hardware, wired connections, no internet involved at all) there’s definitely a hit when it comes to input lag and compression artefacts. Maybe the power of Google can reduce that but I’m not convinced they will ever eliminate it.

Still, Brian always thought OnLive was the future so who are we to argue?

ON LIVE COULD STILL BE THE FUTURE

So could the Gizmondo, but it’s fairly unlikely at this point in time.

Never write off the Gizmondo.

OnLive wasn't great, but it meant that in the middle of nowhere in France I could play Borderlands on a rubbish five year old PC.

There's certainly potential in the concept, all Google need to do is get the teen-agers interested and it'll be huge.

This will never have the same kind of visual fidelity as a proper home console;

Why not? That's certainly what Google are aiming for, and it definitely like like they'll have the hardware for it. The only question mark is your connection speed, and those are getting better all the time.

I played a couple of games on GeForce Now – WoW and Portal 2 – and the experience was flawless. Matched my laptop's resolution perfectly, no artifacts, 60fps, zero (detectable) input lag. Even sat close to a laptop screen, it would have been impossible to tell you were streaming if you didn't know. And that was over wifi.

So the potential is definitely there. Pretty crap presentation though, unless they were aiming solely at investors.

Why not. If it works and it's seamless it's how we'll all consume games fairly shortly. We're already miles down the road away from the thrill of a disk product and manual sniffing.

I'm not advocating boxed product over digital. Far from it. In this case though:

• It's a streaming service, meaning I possess absolutely nothing of what I own. If a game gets pulled down, it's gone. At least now, if it's on my HDD I can still play it or I've bought it, I can still re-download it in most cases even if it's not available to buy.

• The pad is atrocious and connects to WiFi, not the device you're streaming from. Lagtastic.

• Internet infrastructure in many countries isn't good - the UK is still using fucking copper wires, even for things like BT and Sky that say they offer fibre services! - and not everyone can afford/is in range of proper fibre optic stuff. John Smith has 300mb fibre and is hard-wired with ethernet to enable smooth data transfer? Bully for him. I'll sit here with my 40mb wifi router in another room that likes to drop out whenever it feels like it and see how that goes.

• Google has absolutely no clue how to curate a store for games and seemingly no desire to. It's going to be an absolute shitfest of genuinely good AAA titles (which you can play on other services) and utterly dreadful mobile ports, knockoffs and other crap.

• Google's pitch is currently awful and doesn't relate to the average person. I don't give a shit about teraflops or other bollocks, just show me that it works in the wild. Not, as is currently the case with all outlets who've seen it so far, on a Google campus linked to Google's infrastructure running on Google Pixelbook tech that costs a shedload of money.

That's just off the top of my head why it's not for me.

EDIT: Also, everything this guy says (which is a lot of what I've said, but he's more qualified to say it than I am).

The only question mark is your connection speed, and those are getting better all the time.

That's a fallacy. They're only getting better a) if you can afford to pay for it and b) are lucky enough to be near somewhere that supports better tech. As I said, much of the UK is still running on ASDL copper wires - most fibre services are only fibre to the central hubs or the junction boxes, not your front door. Thank Thatcher for that one, I believe. Until there are massive upgrades to tech on a grass roots level in most countries, this is pie-in-the-sky future stuff that just won't be good for a lot of people.

The pad … connects to WiFi, not the device you're streaming from. Lagtastic.

I believe the theory behind that one is that it connects directly (and separately) to the server that's running your game, so you don't have the "console" (Chromecast, laptop, whatever) as a middleman.

One thing I've not seen discussed much is the "watch a video on YouTube and click play now" aspect. Surely that suggests a subscription model rather than buying the games outright?

I believe the theory behind that one is that it connects directly (and separately) to the server that's running your game, so you don't have the "console" (Chromecast, laptop, whatever) as a middleman.

That's sensible for a streaming service, but still comes with its own potential pitfalls. There are currently numbers floating around for the pad latency from Google's GDC stand where Doom and Assassin's Creed are being shown. 107ms isn't exactly a long time, but that's still more than you'd want (though, yes, I'm aware that the XBO pad latency is roughly 145ms on a direct connection). Sadly, that's in a controlled environment where Google has done everything it can to make it the best. No idea what it'll be elsewhere.

We'll see, I guess. No point speculating until it's in the wild and we see how it all works on an average set-up.

One thing I've not seen discussed much is the "watch a video on YouTube and click play now" aspect. Surely that suggests a subscription model rather than buying the games outright?

It could be both, I suppose - if you don't own the game, it could always bring you to a store page to buy it.

What seems to have been completely overlooked by Google is any explanation of the business model or how they plan to attract developers to make games for it. Just saying 'It's the next big thing, you'll want to be on it!' isn't enough if there's no financial incentive or potential to make money. A subscription model just doesn't work for a small/mid-sized studio looking to self-publish titles.

It's another SKU, another platform to consider (assuming it's not running on PC code or whatever, I know it'll be a Unity output plug-in), more cost for developers to create and then support for the future. I strongly suspect the majority of indie devs won't even think about it until it's a proven entity. They just won't be able to afford it.

One thing I've not seen discussed much is the "watch a video on YouTube and click play now" aspect. Surely that suggests a subscription model rather than buying the games outright?

It could be both, I suppose - if you don't own the game, it could always bring you to a store page to buy it.

I'd considered that but I'd assumed a store page would be just another obstacle in the way of getting the game to the player.

It's another SKU, another platform to consider (assuming it's not running on PC code or whatever, I know it'll be a Unity output plug-in)

It's running on some flavour of Linux, though it wouldn't surprise me if it was heavily customised. The Vulkan API – and partnerships with Unreal and Unity to get in-engine support – should make it relatively straightforward for developers to port stuff to Stadia; in the presentation they claimed it only took a few weeks for Ubisoft to convert Assassin's Creed Odyssey.

This was obviously aimed more at AAA publishers than indies, though – and I'm sure Sony, MS and even Valve won't be too pleased by a lot of what they saw. Nobody else – except, maybe, Amazon – has the infrastructure or engineering capacity to build something like this, and not in any kind of timescale that could compete with Google's launch window.

All you need to get started on this thing, as a consumer, is the Chrome browser – no hardware investment required. If playing on the TV is essential, you just need a Chromecast, which is only £30-70. Even if the controller is £50 (around the PS4 pad RRP), that's 4K gaming for about the cost of a Nintendo 3DS.

Of course there's a lot of questions about how it'll perform in the real world – but if anybody has the resources to succeed at this, it's Google.

It's running on some flavour of Linux, though it wouldn't surprise me if it was heavily customised. The Vulkan API – and partnerships with Unreal and Unity to get in-engine support – should make it relatively straightforward for developers to port stuff to Stadia; in the presentation they claimed it only took a few weeks for Ubisoft to convert Assassin's Creed Odyssey.

Ah, yes, Ubisoft… that company famous for not having huge amounts of resources to hand and people it can just throw at things. A few weeks of Ubisoft time is several months minimum for small teams, if not significantly more (and that's time not spent doing anything else). Linux in any flavour is a hellhole from which most indie devs I know steer well clear of. Ugh.

All you need to get started on this thing, as a consumer, is the Chrome browser – no hardware investment required. If playing on the TV is essential, you just need a Chromecast, which is only £30-70.

I have a Chromecast. It's not a fan of the BT router I'm having to use currently and either drops out randomly (network issues, not specifically the Chromecast's fault) or runs images like a dog when streaming things like movies from my phone. So, that in itself isn't something that inspires me when extrapolating my circumstances (we have BT's 'top' fibre package and my router is in the next room) onto
a global average person's scale. Again, it's not entry-level cost that's my concern, it's how well it'll work when I have it.

Of course there's a lot of questions about how it'll perform in the real world – but if anybody has the resources to succeed at this, it's Google.

That suggests that all the problems can be solved by Google throwing money at them. That's not the case at all. They appear to be very much in 'Hey, if we market the shit out of this, it doesn't matter if it doesn't work great for a lot of people!' mode. We'll see when it gets here, I guess.

It's a streaming service, meaning I possess absolutely nothing of what I own. If a game gets pulled down, it's gone. At least now, if it's on my HDD I can still play it or I've bought it, I can still re-download it in most cases even if it's not available to buy.

You don't own games at the moment. Just a licence to play them. Pretty much anything post 360 era can be crippled by the removal of online authentication to a service.

The pad is atrocious and connects to WiFi, not the device you're streaming from. Lagtastic.

This is a good idea - connecting directly to the datacentre should reduce lag.

Internet infrastructure in many countries isn't good - the UK is still using fucking copper wires, even for things like BT and Sky that say they offer fibre services! - and not everyone can afford/is in range of proper fibre optic stuff. John Smith has 300mb fibre and is hard-wired with ethernet to enable smooth data transfer? Bully for him. I'll sit here with my 40mb wifi router in another room that likes to drop out whenever it feels like it and see how that goes.

True - but people will accept lower fidelity for convenience. It doesn't need to work perfectly at 60fps, 4k for everyone. it just needs to work. I know plenty of people who consume Netflix on a 1Meg connection (my partner being one of them).

Google has absolutely no clue how to curate a store for games and seemingly no desire to. It's going to be an absolute shitfest of genuinely good AAA titles (which you can play on other services) and utterly dreadful mobile ports, knockoffs and other crap.

If it gets big (and one of these services will), then they'll buy Steam or something. I doubt they have any desire to run a curated store. They want to run the infrastructure.

Google's pitch is currently awful and doesn't relate to the average person. I don't give a shit about teraflops or other bollocks, just show me that it works in the wild. Not, as is currently the case with all outlets who've seen it so far, on a Google campus linked to Google's infrastructure running on Google Pixelbook tech that costs a shedload of money.

It is primarily a pitch to investors and people with money. The killer feature is the click through from a Youtube video. Get influencers pushing it and it'll explode quickly. Provide a flat monthly service charge and I could see this being enormous.

I understand the skepticism around this (and honestly I do share a lot of it), but I'm not sure anybody knows network infrastructure like Google. If they say they can pull this off in the real world, I'd be pretty sure they have the data to back that up.

Of course, all these things are different when they're in the wild and it might fall over completely. Maybe no developers will really buy into it given the additional development time and unclear (for now) business model. Early impressions from streamers and media are going to be vital for this to take off (I wouldn't be surprised to see a beta period or staged rollout beyond the geographic limitations they're beginning with) and if it's not rock-solid out of the gate, internet hyperbole could kill it dead.

I just wouldn't bet against it.

A few weeks of Ubisoft time is several months minimum for small teams

I doubt Ubi threw an entire dev studio at this, but point taken.

If [Google] say they can pull this off in the real world, I'd be pretty sure they have the data to back that up.

They also thought they could pull off Google+, though.

G+ failed for UX reasons, not the technical infrastructure behind it.

G+ failed for UX reasons, not the technical infrastructure behind it.

Not the fact that it was a thing no-one wanted then?

I call that a User Experience failure, yeah.

Well, the users wanted an experience that didn't involve Google Plus.

I realise I'm playing devil's advocate a lot here (the devil being Google, in this case) but one thing G+ had to contend with that Stadia doesn't is a couple of other big, widely-adopted and high-profile competitors doing broadly the same thing.

Nobody else has a meaningful foothold in the streaming games "space" yet – maybe that's because it's not a product anybody wants, maybe it's just because nobody's really put in the effort (and cash) yet.

I've seen comparisons to VR, though – which I think is an apt comparison in some ways more than others. There's no similarly-expensive hardware required, and the odds of motion sickness and eyesight precluding people from playing seem low with Stadia. But there are significant technical hurdles to overcome, as well as a healthy amount of skepticism from the people they want to target.

In not linked news, but because you mentioned VR, I saw Oculus today say about the new Rift S:

"We really think $399 is a sweet spot for both these products, and we think a lot of people are going to be excited about that price."

(bearing in mind the original Rift was the same price, I think)

and also:

"We plan on making sure that software comes with you, so there's really no reason not to buy now"

HAHAHAHA fucking idiots, they've learnt nothing.

Took shots at the Dreamcast with there little presentation case thingy. Wankers, I'd love smash a wrecking ball through their smug little campus for that.

Google's been streaming modern games for years. It's called "YouTube". In a world where games are 99% story and 1% gameplay, watching a game on YouTube is basically the same as playing it.

Google's been streaming modern games for years. It's called "YouTube". In a world where games are 99% story and 1% gameplay, watching a game on YouTube is basically the same as playing it.

Google's been streaming modern games for years. It's called "YouTube". In a world where games are 99% story and 1% gameplay, watching a game on YouTube is basically the same as playing it.

This is the bit where I can't be bothered to source that GIF of J. Jonah Jameson laughing his head off. Made my Thursday morning at least though, so thanks for that.

Google's been streaming modern games for years. It's called "YouTube". In a world where games are 99% story and 1% gameplay, watching a game on YouTube is basically the same as playing it.

I realise that this facetious non-point doesn't really bear engaging with, but there is at least one major technical difference between streaming video and streaming a live interactive game: you can buffer a video.

I come back the simple proposal to link to Youtube (and one would assume other streaming services) so there is click through play. That's the killer difference between this and other services. A seamless integration of passive game consumption and the occasional dib in to play bits is absolutely how the next generation will want to play longer titles - however Dokky Mc Dok7 place that quote was.

I can imagine sticking on Ass Creed 24 and browsing the Internet while Influencer69xx plays the intro sections for me. Oh, a bit in a boat, I'll do that. Oh god, this bit is hard, let it seamlessly drop back to a streamer. It's some sort of fresh hell, for sure, but it's games in 2035.

I've already made peace with not "owning" things I spent 20-odd years hoarding. Happy to get on board with this assuming it works and the price point suits. But I could also see a case for still owning a box and accessing this service via a more traditional console assuming it held my digital library.
The idea of being involved in a game and then being able to pick up from where I left off on my work laptop while away from home is great. The reality of having to pay Premier Inn £5 to unthrottle the internet connection will still rankle.

I saw an additional thing yesterday saying that, on average, the service will be using 20GB of data an hour. I get that a lot of people are now on 'unlimited' services these days, but a lot of providers still employ throttling techniques for users with excessive consumption. Plus, some people are still on limits if they're low-income or live in certain places (for instance, Australia where the internet still appears to be stuck in the 90s). There's no way I'd use any device not attached to my home connection to consume that much data. No way.

Let's hope Google don't force you to do that at gunpoint.

Sigh.

The point is that they're saying you can play anywhere on pretty much anything. But you can't, can you? Well, you can but the quality/fidelity/lag will suffer greatly unless it's on a network that's good enough. In which case, you can't. That's like saying you can have steak any time, but it'll only taste good at home. Hope you like shoe leather! Etc, etc.

There's an official stream happening later this afternoon (or earlier today, or yesterday, or the other week, depending when you read this) with further details about Google Stadia, but it looks like some of the pricing and other bits have already been leaked.

• There will be a base subscription available when Stadia launches in November, for $11.99 Canadian (£7) per month. This one will give you access to some Stadia games at 4K resolution/60FPS.
• Other games—newer ones, especially—you’ll have to buy.
• In 2020, Google will launch a "Stadia Base" model that’s totally free, but only allows you to play games at 1080p resolution.
• A pricey Stadia Founder’s Edition will be for sale this fall for $169 Canadian (£99) that includes a Chromecast Ultra, a Stadia controller, a three-month subscription, and Destiny 2 (which, as Kotaku revealed yesterday, will be on the platform at launch).
• Doom Eternal, Assassin’s Creed Odyssey, the new Tomb Raider trilogy, and The Division 2 are also part of the Stadia launch line-up.

This is now free for everyone. 2 months of free 'pro', or permanently free basic edition, which gives you 1080p60.

I just gave it a whirl and tbh, it's pretty impressive. Within 60 seconds of clicking 'Try Stadia' I was playing Destiny 2, with my existing Xbox character, on a Macbook that would struggle to run Minesweeper. There's input lag – it's akin to a TV before you've enabled 'game mode'. So it's not perfect. But the tech itself is pretty close.

The only problem is… there's nothing to play. My options are Destiny 2, which is free on every console and I didn't even like, and… that's pretty much it. Unless you fancy paying 55 quid for AC: Odyssey's basic edition. Or 50 quid for Borderlands.. Or 75 quid for Doom Eternal.

I can't believe that Stadia continues to exist in a world where X Cloud is a thing.

I suppose the (potential) benefit of Stadia is being able to play PC exclusives. I already have an Xbox, so don't need X Cloud. Although Stadia has completely dropped the ball by not actually getting any PC exclusives.

Geforce Now is still the most useful of these clients I've found. Access to your existing Steam library is a godsend.

I gave this a blast for ten minutes or so, and was distinctly less impressed with Destiny's performance. perhaps playing with a mouse was a mistake when I'm used to a pad, or maybe my connection's just not as silky-smooth as MPH, but I found the controls quite tough to get used to. There seemed to be some kind of response curve on the mouse that smoothed out/removed small movements, but exaggerated large ones, leading to my view spinning 360° (or more) when I was trying to whip across to a moving enemy.

I had a couple of pretty big visual degradation and lag moments, too, and generally it just felt very floaty and disconnected. But then I've not played Destiny 2 in a good while and maybe it's always been like that. ¯\_(ツ)_/¯

In general I'd find any sort of lag to be much, much more noticeable with a mouse than with a pad.

Yeah I connected my Xbox pad via bluetooth. It would definitely have been annoying with a mouse.

But no I didn't have any technical hitches, aside from that input delay the stream was perfect. I have 100mb fibre, which I don't think is top-of-the-range or anything? And I was using wifi.

MPH's post has twists and turns like a Hollywood blockbuster. No, didn't use a mouse, explaining the latency issues, but… he used a pad with Bluetooth? OK that's weird, but phew it was good that his stream was perfect with his…. 100Meg fibre, which isn't even available where I am… oh hold on he thinks that's not top of the range and woah, he was using WiFi, what the fuck?

saw a gaming laptop that only did wifi internet connections

almost died

This must be highly locational as I can't see any option to try it. Perhaps it's based on proximity to a data centre (and I expect my nearest one is on Mars).