What initially seemed like a really promising idea, to give you gaming on any device and wherever you are has turned into something of a let-down overall.
This will no doubt get me some flak from Stadia stans, but let's keep in mind I was originally totally sold on the idea of Stadia. I have a Founders pack and I used it almost daily for quite some time. That time quickly lessened, and eventually became none at all. I can't imagine I am alone in that either.
At the two year point, what did Google do to celebrate Stadia? Close to nothing. On Reddit the Stadia team went over some numbers we already knew like the amount of games available and a few that added special Stadia features. There was also a sale on their store, along with a reasonable discount on the Stadia Premiere Edition (£19.99, down from £69.99), which you can easily put down to them wanting to get rid of stock since it comes with their older Chromecast Ultra. On the subject of the future, they only gave some vagueness:
- Continuing feature experiments with the goal of making it easier for players to get into games and try Stadia for themselves. We’re still learning from input provided by our community and appreciate all the constructive feedback we receive from you!
- Expanding all categories of games content - not just more games overall, but new types of games that we’ve heard players ask for, including genres like online action games, open world titles, plus free games, trials and demos.
- Bringing Stadia to more devices and making it easier to access, purchase, and play games by yourself or with friends.
No player numbers, no sales numbers, absolutely no show of strength.
Barely any effort to mark two years, unless you count talking very briefly to six (yes, a whole six) customers who picked up the Founders pack. Really pushing the boat out there!
It's hard to be excited or even just a bit interested in a service that Google don't seem to know what to do with. It reportedly missed all their user goals by hundreds of thousands, and they shut down Stadia Games & Entertainment before even giving it any time in the spotlight at all. We were supposed to get first-party games that took advantage of the cloud, to do things you couldn't really do locally and we're likely to never see anything like that on Stadia.
The huge problem is that NVIDIA GeForce NOW and Microsoft's Xbox Cloud Gaming both completely destroy it when it comes to price vs value. Even though GeForce NOW still feels a bit too disconnected, since it relies on whatever launchers games use and all the logins that come with it and Microsoft need to improve the latency / input quality of their offering, Stadia will basically never match up to either on overall value. You've also got Netflix expanding into cloud gaming, and Amazon with Luna. The sharks are circling and Stadia is bleeding in the middle.
When thinking on how Stadia operates, it just really doesn't make sense, especially now with the hot competition. Full price per-game to basically rent your games from Google, with an additional extra monthly sub on top to get 4K and access to a few games per month if you keep that subscription up, to completely disappear if they do shut down the consumer store side of things. When elsewhere you can either pay monthly to access your existing games (GeForce NOW), or pay monthly to access a big library (Xbox / Luna). At least with the other options, you either still have local access or you know you're paying for a more Netflix-like model.
Even Stadia as a service for bigger games has been left in the dust often, with some games leaving patches out for weeks and multiple games released locked to 30FPS. Even developers that are on it don't seem to care enough. Google don't even put Stadia at the front of anything they do, like how their newer Chromecast with Google TV took nearly a year to support Stadia.
Specifically when thinking about the Linux desktop, some original thoughts were that since Stadia was using Debian Linux and the Vulkan API, that we might see some cross-over of ports but that never really materialised either. The majority ended up just sticking to the Stadia ecosystem.
Where does Stadia go from here? Well, we already know they're marketing their tech as a white-label solution to studios outside of the Stadia Store, so that will likely pull in some companies but eventually I do expect the consumer side of Stadia itself to die-off.
Article taken from GamingOnLinux.com.
Arehandoro Nov 24, 2021
Quoting: scaineQuoting: kuhpunktQuoting: scaineQuoting: elmapulfor example, in my country, if i count air tv and cable tv, we had 100 animes broacasted in total across 3~5 decades, there are 10.000 animes on my anime list alone (not to mention its an incomplete list, it dont have the indie non official touhou animes for instance), crunchroll dont have everything, but at least it has 700 animes any given month.
This is a really good point, tbh, and re-enforces why monopolies are bad - if they control what content you watch, you're severely restricted as a focused client of that service. But while dealing with multiple streaming services feels bad, they're still actually really cheap compared to what we used to pay for scheduled television, and they allow a huge degree of control over what you watch.
My primary gripe is really just finding content sometimes. And a minor gripe about how different services look different. And handle end-of-show credits. And how quickly they stream the next episode before I can find the remote control to turn off auto-stream. So frustrating.
Okay, so a few gripes. Still better than a monopoly though.
In Germany there was Sky for soccer/football. Bundesliga and Champions League... all you needed in one small package if you wanted to watch every game.
Then they had to split up the rights, because it was claimed that competition is good for the customer. Then the rights went to Sky for the Saturday games, but the Friday games were on Eurosport. And some games were exclusive to DAZN/Amazon or whatever. Total clusterfuck and more expensive than before.
Yeah, my point was aimed at scheduled content, but in my head I'd excluded sports for some reason. Sports feels really, REALLY overpriced to me. I pay £21/month for all the tv and movies in the world, but I have to pay nearly £30 a month for the F1??
I suppose it's not targeted. I get all the football, all the golf, all that other stuff I'll never watch. I pay that nearly £30/month for pretty much 2 hours of telly a fortnight, which is a shockingly frustrating model. I guess if they did it any other way, we wouldn't have 22 millionaires kicking a ball around a field.
Unfortunately, I think it'd be the other way around. If they were to be targeted, potentially with football package being cheaper, the amount of people paying only for football would increase, where other less common sports would suffer to get revenue from TV royalties.
Quoting: TheSHEEEPGpu sharing is the key.Quoting: MohandevirThe chip shortage is making cloud gaming attractive and there is no forseeable ending to it, yet.Not really.
If a lot of people picked that up, the cloud gaming providers would themselves have to scale up and would be the ones facing the shortage.
They don't need to scale up for every new user, because one single gpu could serve multiple users.
This can be done in different time moments or even sharing them at the same time, it is technically possible.
So in this chip shortage moment, cloud providers have and will have a big advantage over single users; and on a side note this benefits environment too.
Last edited by kokoko3k on 24 November 2021 at 3:52 pm UTC
Arehandoro Nov 24, 2021
Quoting: kokoko3kand on a side note this benefits environment too.
No, it doesn't. Cloud providers need to have their equipment constantly on, with redundancy, capacity for demand surges, UPS systems, industrial cooling... and all this replicated throughout all their data centres to cater audiences around the world.
Last edited by Arehandoro on 25 November 2021 at 8:42 am UTC
Quoting: ArehandoroNo, it doesn't. Cloud providers need to have their equipment constantly on, with redundancy, capacity for demand surges, UPS systems, industrial cooling... and all this replicated throughout all their data centres to cater audiences around the world.
Why would you want to have more PCs running than what the customers of the next say 20 seconds would need? And even if double the customers log in, have them wait for one boot, it's not like it still takes 5 minutes nowadays... I don't have numbers, but usually an industrial solution to satisfy 100 needs is cheaper than 100 individual ones.
Last edited by Eike on 24 November 2021 at 5:35 pm UTC
Mohandevir Nov 24, 2021
Quoting: EikeQuoting: ArehandoroNo, it doesn't. Cloud providers need to have their equipment constantly on, with redundancy, capacity for demand surges, UPS systems, industrial cooling... and all this replicated throughout all their data centres to cater audiences around the world.
Why would you want to have more PCs running than what the customers of the next say 20 seconds would need? And even if double the customers log in, have them wait for one boot, it's not like it still takes 5 minutes nowadays... I don't have numbers, but usually an industrial solution to satisfy 100 needs is cheaper than 100 individual ones.
I don't know if it's applicable to cloud gaming, but where I work, the rule of thumb is one licence for 2 clients and we are talking about simultaneous full time workers... Just a guess, but I suspect it must be less than that with cloud gaming (not all clients are online at the same time an not all clients are used everyday).
Last edited by Mohandevir on 24 November 2021 at 6:27 pm UTC
Mohandevir Nov 24, 2021
Quoting: ArehandoroQuoting: kokoko3kand on a side note this benefits environment too.
No, it doesn't. Cloud providers need to have their equipment constantly on, with redundancy, capacity for demand surges, UPS systems, industrial cooling... and all this replicated throughout all their data centres to cater audiences around the world.
Oups! There is an error in your quote brackets. I'm not the one who wrote that.
Last edited by Mohandevir on 24 November 2021 at 8:25 pm UTC
Quoting: GuestQuoting: BigJI was excited to try Stadia, but the latency of Destiny 2 was unacceptably shitty. I didn’t look into why this would be, but it was enough to turn me off the service for good.
Did you try any other games? I did have a couple of issues with a couple of games when it launched, my favourite being a 30s (yes, thirty second) response time in Serious Sam 3. It varied from game to game, and mostly fixed up before long. Never tried Destiny 2.
Thanks mirv, I'll give it another shot! I didn't try any other games at the time.
Arehandoro Nov 25, 2021
Quoting: EikeQuoting: ArehandoroNo, it doesn't. Cloud providers need to have their equipment constantly on, with redundancy, capacity for demand surges, UPS systems, industrial cooling... and all this replicated throughout all their data centres to cater audiences around the world.
Why would you want to have more PCs running than what the customers of the next say 20 seconds would need? And even if double the customers log in, have them wait for one boot, it's not like it still takes 5 minutes nowadays... I don't have numbers, but usually an industrial solution to satisfy 100 needs is cheaper than 100 individual ones.
For a variety of reasons:
1. Because powering a server on requires more than 20s. The server needs to initialize, load all the components, and the software that later on the customers will use. The only way to provide a seamless experience is having the equipment on. Of course, this is at the data centre level, not the application level. The app should be agnostic to this.
2. Because economically speaking, you wouldn't buy hardware that is going to be turned off. You calculate your usage, customers, etc, and when the total usage going beyond x% you buy more.
3. Data centre equipment is meant to be on. I know of countless cases where a server had been turned on for ages, working fine, and the moment it got shut down for maintenance, never came back to life again.
4. Monitoring and alerting will fire every time one goes up/down, or will require extra work to make sure there are no faux alarms.
On what way is cheaper? For me to have a 16cpu, 64GB RAM server* nothing is cheaper than owning it myself. It's also powered on only when I need it, saving energy too.
* By server, in my case, anything with those requirements will do. Nothing to do with the monstrosities ran in the cloud like EPYC or Threadripper.
Quoting: ArehandoroFor a variety of reasons:
This doesn't sound convincing to me. (E.g. having a computer that you need from time to time switched off is cheaper than having it switched on, and monitoring should absolutely be able to cope with intentional switching.) But I'm "only" software developer and do not work on server farms.
But either way there's something we missed: For cloud gaming as we would do it, on a PC, both systems have to run, the local and the cloud system. I can't imagine this to win ecologically against the local-only solution. (Gaming on say a mobile might differ again.)
Patreon
PayPal
See more from me