> It's only been a failure because of the lack of games

Which, to be frank, is the one reason someone would use Stadia - to play games.

I get what you are saying, games are essential if you are building a gaming platform

But a lot of people expected it would fail due to network issues, lag/delay, etc. If the tech was terrible, the good games wouldn't have been enough to save the project.

I would argue that the project never got far enough to tell if it was a successful gaming platform since there are no games that people want to play. I guess given the very limited functionality it’s technically not a failure. But the whole point is to have games people want to play.

Compare to Sony PlayStation 1 (which was a phenomenal success):

Since Sony had no experience in game development, the company turned to third-party game developers. Recently released consoles like the Atari Jaguar and 3DO suffered underwhelming sales due to a lack of developer support, prompting Sony to redouble their efforts in gaining the endorsement of arcade-savvy developers.[22] With initial support from Namco, Konami, and Williams Entertainment,[41] as well as 250 other development teams in Japan alone, the company secured the launch of influential new games such as Ridge Racer and Mortal Kombat 3.[22][5]

> the project never got far enough to tell if it was a successful gaming platform

The project hasn't gotten far enough to tell yet. Presumably shutting down their internal studio will help them focus on the platform rather than on making games. Much like Sony, it seems like this would have been the better strategy to start with.

Succeeding here might require both.

Did they expect to make money instantly without massive investment? Seems very shortsighted.

Given that they didn't launch a single game, I think SGE just regular old failed.

In top of that, after seeing Cyberpunk 2077 get the closest thing to an exclusive launch they could possibly hope for, I imagine they looked at the results decided that the problem is not going to be solved by one or two great games that only work on Stadia. They need many games and for that they need third party.

I remember a time where new consoles would launch with like 10 games available. It was fine, more games came later.

> tech (at least to me) feels fine [...] it _could_ be a success but the business side is broken

... said no one about any Google side project before, ever.

Careful, this much sarcasm might break the website

Compare to https://www.nvidia.com/en-us/geforce-now/ and https://www.xbox.com/en-US/xbox-game-pass/cloud-gaming

Both have their own quirks, but they don't suffer from lack of games.

Man, I'm so salty about GFN - they have a great product and an economic model that makes sense but, for whatever reason, they gave in to publishers' pressure to remove some games from their service. This is doubly upsetting, as publishers should have no say on where I play the games I've already purchased - whether it's on a computer I physically own or on a computer I rent in the cloud.

They didn't gave in.

Publishers changed their licence agreement to exclude streaming or cloud.

Since you are free to change the licence at anytime you like, they did. My bet is we will see white labelled streaming services from Blizzard and co.

We will have dozens of gaming streaming services, much like film streaming services...

Sadly, you generally don't purchase a game, but a license to play one in some form. I think the exact limits of their control have not yet been tested in court, but I don't think NVIDIA wants to be the one testing it.

And any way, NVIDIA is in no position to challenge the games industry on such things - they are very much in bed with them and would lose enormously if games suddenly all went for a big 'best on AMD' logo.

Geforce Now library kinda sucks to me, not all that much better than Stadia but that's my opinion. Stadia has a few names (e.g. Red Dead Redemption) that GFN doesn't and vice versa.

It used to be much better, but publishers cracked down on it when it started gaining traction.

I think it's ridiculous that a publisher can prevent me from using my license on what is essentially rented cloud hardware. IP laws have come to stifle the very innovation they were supposed to protect.

> I think it's ridiculous that a publisher can prevent me from using my license

I'm still not over the fact we are all buying licences instead of copies.

Legally it's far from clear they can, but Nvidia entered a market where they can't afford to piss off game makers from their legacy business and folded on it.

This is probably what prevented the Stadia team from doing the same thing. Lot‘s of people saying: „why not partner with valve/steam for their library“. Because it‘s not Valves library to begin with. This was inevitable. No publisher cared about GFN playing their games in the cloud in the beginning. Just like studios sold off their content to Netflix for pennies in the beginning. It was free money and they didn‘t have a use for them anyways.

Now they smell money and opportunity to fleece a bigger company for higher fees. It happens all the time, unfortunately.

That‘s why it‘s pretty disappointing that stadia killed their first party studios. You have to produce your own content (like netflix) or at least partner with publishers to be exclusive or simple. This seems to be happening with Ubisoft.

But MS just has such a massive advantage with Gamepass.

So, I'm wondering, why doesn't google supply us with a desktop in the cloud, you can do whatever you want with it, but it has a nice graphics card so you can install steam on it if you want or origin or whatever. You download the game there and play that way. Publishers can't stop that can they?

Well, it would still be linux presumably. MS would likely not allow them to use Windows that way. Or they‘d make it prohibitively expensive. In any case there‘d be even less differentiation from xcloud.

Shadow does offer Windows VM‘s. It works pretty well, but it‘s also quite expensive and not really generally available.

GFN offers something unique: you can connect to existing stores like steam and play many of the games you already own there over GFN. This is a big deal.

Don't forget https://www.playstation.com/en-us/ps-now/

Still a lot of the big games missing on GFN though, e.g. all of Bethesda and Blizzard games. Apart from that -- a lovely service.

They'll both let you play games you already paid for but Stadia makes you buy them again, unless that changed?

To me, that put Stadia in an odd place where only the more casual players without a gaming device would sign up, or those with enough money to pay twice because they spend a lot of time away (but with good internet).

> It's only been a failure because of the lack of games. T

... which makes shutting down gaming studio part of it even weirder.

Right... I thought as an idea, it was perfectly sound. The types of technology required to make cloud gaming work seemed to play to Google's strengths.

They just really screwed up on the content side. And, well, anecdotally I've heard the tech isn't so great either compared to Google's peers—no idea how that happened, really.

Google had to solve same problem any new traditional gaming console has to solve -- get publishers interested in creating for that console.

Good luck with that. The moat is so high that even if you have tech from year 2222, the business side trumps technology.

Architecturally, are Google's Stadia servers substantially different from a standard PC?

They run linux and use Vulkan for rendering, so this is a potentially large port burden to force on devs.

Potentially, but there's only one distro and only one set of hardware, so there's waaaay less QA than an actual Linux port.

Still, nobody ports to Stadium unless they're paid upfront.

Ah, always wonderful seeing a multi billion dollar company leverage OSS in an attempt to launch a negative sum product as an internal political promotion move only to kill it off a year later. Such shall be recorded in the history of the Net.

Google isn't the company I would lodge this complaint at. They contribute a lot of code back as well.

Google is a company large enough that you can always find fringes that contradict their main philosophies a business model within the same company. They do not contribute proportionally to their revenue nearly as much as the other FAANG companies.

Yes, they do some OSS work; some of it is even good quality; most of it is not, and is open sourced only to increase market grab (looking at you, Android). They are just a mess. We can all cherry pick counter examples, because of how messy they are.

AFAIK they run Linux, so porting games over isn't necessarily "free"

If anyone is going to compete with Microsoft, of all companies, I'd expect it to be Google.

Except Microsoft is investing heavily in first party games, buying loads of studios and letting them retain a lot of independence. They’re approaching the streaming service as an extension of the Xbox experience. It’s a much more robust strategy all around

> and letting them retain a lot of independence

Except they misteriously stop releasing Linux versions of their games.

But there's no reason Google couldn't have done these things! They just... didn't.

And not forcing studios to rewrite their games from scratch to a snowflake platform.

Porting a game to Linux and Vulkan does not require rewriting it from scratch.

It does because 99% of the game studios don't care about Linux, nor are that much into Vulkan, unless they are targeting Android or Switch.

The graphics backend is really not such a big cost for most games, especially with existing cross-platform engines. Also the grahics APIs are really not that different as they all model the same GPUs. DirectX vs. OpenGL had some annoying things but most of them are only a problem when you have to translate and can't modify the calling code at all. DirectX 12 vs. Vulkan is even less of a difference.

How many Android and Switch games get published on GNU/Linux? Yep, it boils down to it.

Let's cool it with the hyperbole, okay?

There are many reasons Android and Switch games don't get ported to Linux; having to rewrite the game from scratch is not one of them.

Which proves the point that even when games happen to be natively written against a Linux userland, with native support for Vulkan, it isn't economically viable to sell them on GNU/Linux.

What's the alternative? Run Stadia on Windows?

That is exactly what Microsoft is doing.