Google Stadia - future of gaming?

Has anyone heard about this before? If someone posted it here I missed it (feel free to delete the thread if there is one).

Not sure what to make of it. The future of gaming?

https://www.gameinformer.com/gdc-20...-stadia-a-powerful-new-game-streaming-service

What Games Can I Play On Stadia?

Hundreds of publishers and studios are currently migrating games to Stadia. Bethesda, Ubisoft, Take-Two, Q-Games, and Tequila Works all participated in the presentation.

Click to expand...

Yea, we were under impression VR helmets are the future of gaming and you know how that ended.

It's still unclear what's really going on there, a customer will want smooth experience without SATA2 endless loading screens for example and I'm not so sure even google can make it happen for hundreds of thousands people playing games at the same time. We'll see.

Oh I'm sure they can make that happen. The question is, is it profitable (for google) and affordable (for gamers).

Once the ISP infrastructure can support it reliably, I would think there's a pretty good case for it. If you think of it as massive data centres full of racks of mini consoles, the hardware will likely be utilised pretty consistently by different users. A typical piece of gaming hardware probably sits idle for the majority of the time. I think the economics of it would look pretty good.

Not keen on the idea myself, but I could see it coming eventually.

While Gameboy could provide hairworks that way, there is still something the service won't be able to cover.
Mods.

joxer said:

While Gameboy could provide hairworks that way, there is still something the service won't be able to cover.
Mods.

Click to expand...

It could be done. People could upload mods to their Google Drive or somesuch, and then have the mods applied temporarily to their instance when it fires up. Brings up security issues on their end, but not insurmountable.

Unless you'll have to pay to install inside google app store... Never gonna happen.

Yeah, i think they'd be inclined to have their own paid and vetted mod store, rather than letting users apply their own mods, but it's not necessarily a restriction of the technology.

Never heard of this before and I'd be very surprised if it ever turns out to be something huge.

It must add some slight latency, even on a fast connection, even if it's in the 40-50ms range i probably wouldn't want that, because these latencies you already have (screen, input device) it does add up and it will be noticeable.

It's just streaming again. There's a ton of very nice things about streaming and there's a ton of really bad things, too.

• After you buy a game, you play it immediately. No download, no install, and certainly no having to drive to a store. {This makes demos awesome! You could pop through one after the other after the other until you find a game you like.}
• Mods will be difficult at best - especially ones with copyright issues.
• For fast action games/fighters, lag will be a thing. For turn based, it will be totally irrelevant.
• Death to pirates! Also turns the 'used game' market into nonsense. Publishers are going to love this - probably to the point of helping fund these sorts of services. It will be a positive for folks like me that don't pirate and rarely buy used, too, as more people will be paying.
• Fancy display hardware dies, too. VR and AR are going to be far down the list of priorities. I'm not sure about fancy input. I would hope joysticks work but IR head tracking?
• It isn't just that everyone has access to high end PC hardware, it's that everyone has access to the SAME high end PC hardware. Heck, the developers can specify exactly what drivers and codecs are on the server! No more worries on our end about mysterious lock-ups, no more worries on the developers' end trying to support a whole range of drivers, hardware, and heaven-only-knows-what plugged into USB ports.

joxer said:

Yea, we were under impression VR helmets are the future of gaming and you know how that ended.

Click to expand...

<ahem> https://www.pcgamer.com/hps-new-mixed-reality-headset-has-almost-twice-the-resolution-as-a-vive-pro/

Where did I hear this before about cloud gaming being the future? Oh that's right it was a service called Onlive. It failed after years of struggling till it was sold cheaply.

Now it's a defunct company that no one ever talks about.

Zloth said:

It's just streaming again. There's a ton of very nice things about streaming and there's a ton of really bad things, too.

• After you buy a game, you play it immediately. No download, no install, and certainly no having to drive to a store. {This makes demos awesome! You could pop through one after the other after the other until you find a game you like.}
• Mods will be difficult at best - especially ones with copyright issues.
• For fast action games/fighters, lag will be a thing. For turn based, it will be totally irrelevant.
• Death to pirates! Also turns the 'used game' market into nonsense. Publishers are going to love this - probably to the point of helping fund these sorts of services. It will be a positive for folks like me that don't pirate and rarely buy used, too, as more people will be paying.
• Fancy display hardware dies, too. VR and AR are going to be far down the list of priorities. I'm not sure about fancy input. I would hope joysticks work but IR head tracking?
• It isn't just that everyone has access to high end PC hardware, it's that everyone has access to the SAME high end PC hardware. Heck, the developers can specify exactly what drivers and codecs are on the server! No more worries on our end about mysterious lock-ups, no more worries on the developers' end trying to support a whole range of drivers, hardware, and heaven-only-knows-what plugged into USB ports.

<ahem> https://www.pcgamer.com/hps-new-mixed-reality-headset-has-almost-twice-the-resolution-as-a-vive-pro/

Click to expand...

Some fine points there, but I still prefer to have the game locally and do what ever I'm pleased with it.

Stop showing your ignorance. SATA2 is not the major factor esp when most consoles still use hard drives.
-
Anyway the biggest issue is likely latency. For a regional data-center you are likely talking about 5 to 15ms latency; for something further you might see 20ms to 50ms with 25ms probably the average worse case. Lets say the majority of people have a regional data center and the latency is 8ms to 10ms (this ignore the actual processing/machine latency but you should assume that at least initially it will be fairly fast.
-
Note that many tvs have > 10ms lag so while the lag is additive it isn't outrageous but is measurable. This gives you an initial data point that you can use to compare with your hardware and gaming style. For example an older lcd monitor might have 20ms latency and a fast gaming monitor might be 1 to 5ms.
-
I would be more concern about the lag induced by having an application like the browser perform the rendering - though I suppose google will optimize chrome for this sort of thing with the launch of the product.
-
As for cost - this is likely a win for google. I don't have data on google servers but I suspect they figure gamers are active after business hours and that they see more idle times on their servers after business hours so they hope to suck up that idle time with something profitable. Hum. I should get some real data here.
-
Now if the product picks up there might become a point will it is less profitable (induces significant data center cost) but I guess it also depends on what type of games people play. A site like this one is probably does not represent the general population but if a lot of people are playing candy crush and sim (ea is not likely to allow google to actually sell sim but that is a tangent issue)…..

joxer said:

Yea, we were under impression VR helmets are the future of gaming and you know how that ended.

It's still unclear what's really going on there, a customer will want smooth experience without SATA2 endless loading screens for example and I'm not so sure even google can make it happen for hundreds of thousands people playing games at the same time. We'll see.

Click to expand...

you said:

As for cost - this is likely a win for google. I don't have data on google servers but I suspect they figure gamers are active after business hours and that they see more idle times on their servers after business hours so they hope to suck up that idle time with something profitable. Hum. I should get some real data here.

Click to expand...

AFAIK, they won't be using their general servers for the task, which would likely not be suitable. They are using racks full of custom built APUs. So, there won't be any additional efficiency from utilising slack in the general servers.

Interestingly, they've also announced that these systems will be running Linux and Vulkan - which means all these AAA games they're touting are being ported to Linux. Whether we might see them outside this context is another matter, of course.

Yea i saw that but I suspect the servers will be shared. Most of hte internet companies have fairly good load balancers and if the game servers are idle they can be used for general cpu tasks - as their (google) software runs exclusively on linux.

Ripper said:

AFAIK, they won't be using their general servers for the task, which would likely not be suitable. They are using racks full of custom built APUs. So, there won't be any additional efficiency from utilising slack in the general servers.

Interestingly, they've also announced that these systems will be running Linux and Vulkan - which means all these AAA games they're touting are being ported to Linux. Whether we might see them outside this context is another matter, of course.

Click to expand...

you said:

Yea i saw that but I suspect the servers will be shared. Most of hte internet companies have fairly good load balances and if the game servers are idle they can be used for general cpu tasks - as their (google) software runs exclusively on linux.

Click to expand...

Yes, that's true. It could run the other way - with the game APUs taking up general tasks when not in use for gaming.

There are several companies that use gpu for encryption. The APU would just serve the same function but might be easier to program. Anyway it will be amusing to see how the service perform. I think hardcore fps players would reject it due to latency but that is likely a minority of the market. The TV folks will likely hop on this (but i could be mistaken). They won't have to mess with a pc at all - just buy the controller
-
Of course none of this is relevant to myself - just curious to see how it pans out. I could see it being a big success or total bust - just not in touch enough with mainstream to know which way they will go - after all i would never had predicted the huge mobile market (huge in $$$ not people count - i expected the people count but i would never pay to play games on my tablet/phone).

Ripper said:

Yes, that's true. It could run the other way - with the game APUs taking up general tasks when not in use for gaming.

Click to expand...

Seems like something that would be more of a scare to the consoles. They don't mod games anyway so the thing that (at least to me) seems like the worst part of this isn't relevant to them. What will be relevant is "no used game sales," though. Microsoft tried to do that and got squashed hard. But what if prices are such that gamers are already paying used game prices right off the bat?

Speaking of price, there's no word on that yet and there's a LOT of ways this could go. (List time again!)

• Pay $X and play the game as long as you like.
• Pay some small amount per minute.
• Pay per minute but the first hour (or whatever) is free.
• Rent to own: just as the previous but, once you've paid $X, the per minute charge goes away.
• Subscription: pay $20/mo, play any game in the store as long as you like. (Or maybe just games that have been there for 6+ months.)

Combinations of the above are possible, too. Heaven help whatever groups in the service companies that are trying to figure out just what one people & publishers will like that can actually turn a profit.

Knowing google I would suspect one of two models; monthly subscription probably unlimited but map some cap (like cable internet); or per game price (library approach). Will be amusing to see what they decide. I'm sure they done some modeling on the backend cost for various models based off of typical consumer behavior. One negative is they can't take advantage of east/west (global and probably local) due to latency issues. Pity they can't improve the speed of light ....

Zloth said:

Seems like something that would be more of a scare to the consoles. They don't mod games anyway so the thing that (at least to me) seems like the worst part of this isn't relevant to them. What will be relevant is "no used game sales," though. Microsoft tried to do that and got squashed hard. But what if prices are such that gamers are already paying used game prices right off the bat?

Speaking of price, there's no word on that yet and there's a LOT of ways this could go. (List time again!)

• Pay $X and play the game as long as you like.
• Pay some small amount per minute.
• Pay per minute but the first hour (or whatever) is free.
• Rent to own: just as the previous but, once you've paid $X, the per minute charge goes away.
• Subscription: pay $20/mo, play any game in the store as long as you like. (Or maybe just games that have been there for 6+ months.)

Combinations of the above are possible, too. Heaven help whatever groups in the service companies that are trying to figure out just what one people & publishers will like that can actually turn a profit.

Click to expand...