Cloud gaming - the end of the desktop performance pc cpu market?

[Idontcare]

Let's ignore the whole latency issue and pretend that that is acceptible. At that point, it's the VDI or App presentation model, but with a workload that requires special equipment (GPU) that doesn't allow for resource sharing on that particular piece of equipment. That puts it in a realm even worse than VDI, and regardless of what you may hear, no one is ever implementing VDI for cost savings. Sure, you only have to deal with peak load, but the gear a lot more expensive than home systems.

Now let's talk processing, delivery, etc.

The netflix analogy really doesn't fly. Netflix is purely a bandwidth problem. Given enough bandwidth, and even marginal latency, you can watch a movie.

Remote gaming is much more involved. You have rendering, *then* you have to be able to encode the video. Only at that point can you send it (all Netflix videos are pre-encoded for minimum bandwidth needed). Once the data is there, the remote system has to decode it, display it, and only then can a player act upon it. I've seen onlive before, and the visual fidelity was pretty much garbage in order to be able to encode fast enough to keep latency low enough that it didn't totally destroy the experience.

Now, the speed of light isn't going to change, but the encoding time will, over time, drop. The decoding time will as well, though they'll never be 0. While there is room for improvement, there is no magic sauce that will remove the latency entirely, and as we're trending towards (slowly) higher resolution displays, the encoding improvements may or may not be entirely offset by having to drive higher expected resolutions.

Netflix is easy. Pre-encode in the most efficient but reasonable to decode format, and add enough bandwith. That just doesn't translate in to an interactive medium.

If everybody on the planet knew how to make it happen then it would have already happened. There is a reason Netflix came into existence instead of hundreds of simultaneous business -netflix knew what to do at the right time for the technology.

I don't claim to know any of the answers, but I am pretty sure that if this were 1980 there would be no shortage of people saying the idea of streaming movies was impossible because the bandwidth necessary would exceed that of all the global governments combined at the time.

Sure we can say now, looking back, that netflix is possible because bandwidth was the answer. But it wasn't us that made that bandwidth possible, it was thousands of much smarter engineers who created the mathematics that were necessary for the protocols and the technology that has come into existence to make that bandwidth possible in the first place.

It doesn't surprise me that we can all come up with reasons why cloud gaming is currently not a reality...none of us are likely to be in any of the meeting rooms, rubbing elbows with the people who are busy creating whatever it is that we will all be buying for $100/month in 20yrs time.

I have a lot of confidence in those guys. They've done some wonderfully impossible things in my time.

 

[Gryz]

We (the readers of this website) are looking at problems from a technical perspective. What architecture would make the most sense, be most effective, lowest cost, best quality, etc.

It is my believe that technologies like OnLive are not developed from the same perspective as ours.

The big benefit of OnLive is not for the customers.
The big benefit of OnLive is for corporations.

Game publishers will keep a lot more control over their product when they sell it via OnLive than when they sell an old-fashioned game that runs on our computers. Of course the risk of piracy is much smaller (approaching zero). Secondly, the publisher now has control over the amount of hours you play, or the amount of times your replay a game. There is no second-hand market. All these properties of streaming games are heaven for game publishers. They will try and keep trying to push this model.



I agree with Idontcare that technology will improve every year. And thus that architectures will change too. But I think there are 2 factors here:
1) CPUs and GPUs become faster, we will get more bandwidth, but the speed of light will not change (see RFC1925).
2) Technology at home, and technology in the data-center will improve at the same speed. Even if technology in the data-center becomes so fast and so cheap that streaming (high-res) games becomes viable, I think that rendering those games on a local device will always be cheaper and more economical.

 

[imagoon]

If the ping is that long then yes, no question, the model itself won't take off for those games which are dependent on latency (FPS). But there is a large group of people who don't need that - they play farmville already.

My point is more to say that obviously the solution here is quantum entanglement as that negates space-time latency at its core. Er, wait, that's not what I meant, what I meant was that they will need to put the servers closer to their customers to ensure that the ping is not deleterious.

I prefer subspace communications 10ms ping even on Risa! =D

 

[kelco]

We (the readers of this website) are looking at problems from a technical perspective. What architecture would make the most sense, be most effective, lowest cost, best quality, etc.

It is my believe that technologies like OnLive are not developed from the same perspective as ours.

The big benefit of OnLive is not for the customers.
The big benefit of OnLive is for corporations.

Game publishers will keep a lot more control over their product when they sell it via OnLive than when they sell an old-fashioned game that runs on our computers. Of course the risk of piracy is much smaller (approaching zero). Secondly, the publisher now has control over the amount of hours you play, or the amount of times your replay a game. There is no second-hand market. All these properties of streaming games are heaven for game publishers. They will try and keep trying to push this model.



I agree with Idontcare that technology will improve every year. And thus that architectures will change too. But I think there are 2 factors here:
1) Speed of light will not change (see RFC1925).
2) Technology at home, and technology in the data-center will improve at the same speed. Even if technology in the data-center becomes so fast and so cheap that streaming (high-res) games becomes viable, I think that rendering those games on a local device will always be cheaper and more economical.

I think the paradigm of desktop rendering and cloud rendering will not be similar.

 

[Gryz]

I think the paradigm of desktop rendering and cloud rendering will not be similar.

I have no idea what you mean by that.

 

[imagoon]

If everybody on the planet knew how to make it happen then it would have already happened. There is a reason Netflix came into existence instead of hundreds of simultaneous business -netflix knew what to do at the right time for the technology.

I don't claim to know any of the answers, but I am pretty sure that if this were 1980 there would be no shortage of people saying the idea of streaming movies was impossible because the bandwidth necessary would exceed that of all the global governments combined at the time.

Sure we can say now, looking back, that netflix is possible because bandwidth was the answer. But it wasn't us that made that bandwidth possible, it was thousands of much smarter engineers who created the mathematics that were necessary for the protocols and the technology that has come into existence to make that bandwidth possible in the first place.

It doesn't surprise me that we can all come up with reasons why cloud gaming is currently not a reality...none of us are likely to be in any of the meeting rooms, rubbing elbows with the people who are busy creating whatever it is that we will all be buying for $100/month in 20yrs time.

I have a lot of confidence in those guys. They've done some wonderfully impossible things in my time.

I believe that the experience could have the potential to become "decent" don't get me wrong. I think the main issue I see is still the physics issue. [right now.] Even if we got bandwidth to be unlimited and the servers "close" we still have simple propagation issues. I used light in this case because it is a the best case we have at the moment. Propagation of electricity through a conductor is even slower. It will take something like quantum entanglement or "subspace" or something else that some how sides steps that issue before we can reduce the latency even remotely close to running most games on $500 desktops.

The main issue here also tend to be where the latency is. The problem is that latency in the video stream tends to be more perceptual to most people vs say 300+ ms server communications on Diablo 3. I would need to go search for sources but some of the original issues with the VR helmets was that even a 20ms delay is enough to start messing with the brain in a way that can cause nausea and the like.

http://www.extremetech.com/gaming/1...world-finally-ready-for-virtual-reality-games

Motion sickness is caused by a disagreement between your vestibular nervous system and your eyeballs. The vestibular system is responsible for maintaining your balance and spatial orientation.

I am one of the people that really can't watch 3d movies for long because I get headaches. While this isn't apples to apples, the effects are related.

I think in the near future the "mmo" style or the "always on diablo 3" style will be the most common because they can hide the latency behind a client that adapt for it.

 

[imagoon]

I have no idea what you mean by that.

I think the OP is starting to troll his own thread. It sucks because we have pretty good discussion going. More "highly technical" than 1/2 stuff in HT as of late.

 

[WhoBeDaPlaya]

I am not a fan of cloud gaming. Look at what has happened with Diablo III. Is a company out there that has more experience then Blizzard on rolling out patches and keeping the customer connected to their game environment? But no they have dropped the ball so many times and still there is hacks other things that must be dealt with before any type of cloud gaming can become real.

Always online is more secure my big blue a*s :|

But don't fall victim to this "technology as a snapshot in time" mentality. Bandwidth is improving moving at ridiculous speeds. Hell, the very fact that the concept of bandwidth itself is prevalent enough that you and I can even have this conversation about the very topic is absurd proof of the pudding type stuff.

I'll believe you when I can get FIOS in Addison/FB. I don't think my employer would take too kindly to me gaming using the corporate pipe, even though it is so thick and juicy

I think the paradigm of desktop rendering and cloud rendering will not be similar.

No offence kelco, but you kinda sound like SCIgen :
http://apps.pdos.lcs.mit.edu/scicache/132/scimakelatex.37095.kelco.html

 

[ShintaiDK]

While we've certainly lost the Internet Speed edge we had 15 years ago - i'd say your perspective is a bit harsh.

The monoply of a certain company is the only thing keeping us back .


I find it funny you use an argument like a single CARD cannot service multiple users.
NVidia is movingly heavily towards more threaded architecture on a hardware level - then it's just the same problem on the cpu-side.
Uber Fast "long distance" interconnects.


Anyhow, the biggest issue is latency and correction for packet loss.
We're just NOT there in terms technology to keep this table in a scaled enviroment (imaging a gaming stream service ala CDN's around the world) - that includes as much upstream to the Server enviroment without speed issues - when comparing to a local Desktop.

We might be at an acceptable level as seen by the OnLive companies and so on - for casual players, but whenever something launches like this it's going to have to be "community\pr" accepted by us geeks first to gain mainstream traction.

That won't happen until latency issues are 99% of a local setup.

Im not harsh at all, its a simple reality. The blame is actually not at that company, but somewhere else.

I think you need to check how graphic cards works. And then try reevaluate how you gonna fit multiple games onto a GPU or series of GPUs.
The nature of games themselves destroy that idea. Its nothing to do with threads. Its about the processing power of GPUs and the memory+bandwidth needed for textures etc.

And I never claimed a single card cant handle more than 1 user. No need to make things up. Try to think on how many modern games 1 card can handle memory and bandwidth wise. Performance is a whole issue of its own. How would Borderlands 2 for example run with 512MB on a 64bit bus?

There is no magic fix to latency. Its a physical limitation until we can send data faster than light. Not to mention to even make something viable economicly. You need to have large consolidated datacenters that again kills the idea. Cloud gaming is a pipedream.

Visually cloud gaming is also...ugly. Just think on the bandwidth needed for bluray quality.

 

[Ferzerp]

Visually cloud gaming is also...ugly. Just think on the bandwidth needed for bluray quality

It's not as much the bandwidth as the real-time encoding requirement that really decides the quality right now, or at least that needs to.

 

[ShintaiDK]

Technology at home, and technology in the data-center will improve at the same speed. Even if technology in the data-center becomes so fast and so cheap that streaming (high-res) games becomes viable, I think that rendering those games on a local device will always be cheaper and more economical.

Exactly. Not to mention all the overhead a datacenter got. Support staff, cooling, extra cost for land/building, backup power, data backup, licenses.

The client PC will still need Windows, CPU, GPU and so on. Nobody in the right mind would pay more for worse quality and experience.

OnLive was a multiple failure in all aspects.

 

[kelco]

I think the OP is starting to troll his own thread. It sucks because we have pretty good discussion going. More "highly technical" than 1/2 stuff in HT as of late.

www.dictionary.com? Thats a pretty clear sentence I formed. I'm postulating that the way all this is computed could very well be completely different than how we would normally do so on a desktop environment.

 

[Rifter]

I thought the same thing of Netlfix. No really! For years I was an ardent and staunch member of the "netflix is never going to fly, who wants postage stamp movies delivered on 28.8k modem speeds?"

And years went by, and I held to my preconceived notions that the very concept of a netflix was dead on arrival.

Then one year while on vacation I was visiting my sister and her husband, and we were watching what I thought was a TV episode of mythbusters. Only when the episode ended I realized it was netflix streaming through a Wii. I had no idea it wasn't actual television/cable until then.

And now I am a happy Netflix subscriber.

I can understand that arguments of why the model of cloud gaming is a no-go in today's world, so was netflix in 1999 when modems were still the dominant bandwidth connection for home users.

But don't fall victim to this "technology as a snapshot in time" mentality. Bandwidth is improving moving at ridiculous speeds. Hell, the very fact that the concept of bandwidth itself is prevalent enough that you and I can even have this conversation about the very topic is absurd proof of the pudding type stuff.

Our parents could not have even dreamed of us having this opportunity, just as we cannot hope to dream of the reality that the future holds for our kids and for us aging enthusiasts.

Never say never, time will make a fool of you every chance it can get

You see thats just the issue bandwidth is moving backwards not forewards.

2 years ago i had no data cap and could have streamed the whole internet with no issues. Now i have a 50GB a month cap and i have to pay more for that over the basic 25GB cap. And there is literally zero local options that are uncapped anymore unless i want a super slow DSL connection that is 1/10th as fast as what i have now.

 

[ShintaiDK]

It's not as much the bandwidth as the real-time encoding requirement that really decides the quality right now, or at least that needs to.

I remember the quality of OnLive was mentioned to be similar to consoles playing on a 480p TV, tho with an extra side blur that often made people dizzy. And with a framerate around 10-20fps.

 

[kelco]

SO the general consensus is that this won't work on the internet. So what about an intranet? Would this lower the cost of owning and operating an Arcade?
Or mitigate the costs of having a lan party business...As we all know most cyber cafes tank which is a damn shame.

 

[ShintaiDK]

SO the general consensus is that this won't work on the internet. So what about an intranet? Would this lower the cost of owning and operating an Arcade?

I dont see how. It would make it more costly as I see it since you need alot more equipment, more licenses, more power consumption, more cooling, more expensive staff to handle the higher technical problems. Network infrastruture to all units. And the famous single point of failure.

Just look at cloud services like EC2, o365 etc. They have alot of downtime.

 

[kelco]

I dont see how. It would make it more costly as I see it since you need alot more equipment, more licenses, more power consumption, more cooling, more expensive staff to handle the higher technical problems. Network infrastruture to all units. And the famous single point of failure.

Just look at cloud services like EC2, o365 etc. They have alot of downtime.

So you can't imagine a mame like system that streams to thin-client arcade machines over cat6?

 

[ShintaiDK]

So you can't imagine a mame like system that streams to thin-client arcade machines over cat6?

The answers is in my previous post. So...why you ask again?

You dont do cloud just to do cloud....

 

[kelco]

The answers is in my previous post. So...why you ask again?

You dont do cloud just to do cloud....

I just wanted to make sure we were on the same page. It sounds like a great way to keep an Arcade current without having to sell an entire game cabinet that no one wants anymore because its not the latest.

 

[ShintaiDK]

I just wanted to make sure we were on the same page. It sounds like a great way to keep an Arcade current without having to sell an entire game cabinet that no one wants anymore because its not the latest.

Not sure if you are serious?

Arcade machines dont exactly use highend hardware. So the cost difference between a thin client and a fixed setup would be minimal at best.

Secondly, all arcade machines got a physical visual design to a game and certain bottons/controls and numbers of players to match.

Thirdly, arcade games almost last forever

Else you could just go to some internet cafe.

I think we reached a point where you _want_ cloud gaming. Try OnLive...oh wait...

 

[kelco]

Not sure if you are serious?

Arcade machines dont exactly use highend hardware. So the cost difference between a thin client and a fixed setup would be minimal at best.

Secondly, all arcade machines got a physical visual design to a game and certain bottons/controls and numbers of players to match.

Thirdly, arcade games almost last forever

Else you could just go to some internet cafe.

I think we reached a point where you _want_ cloud gaming. Try OnLive...oh wait...

Wanting something and foreseeing its inevitability are two different things. I've worked a gameroom, I've been an It administrator and I'm a gamer. Depending on what hat I'm wearing will determine what I'm looking for from a given technology. But its obvious for many reasons why you'd want virtualization in some scenarios. I'll just get back to talking to myself about such things, pretty much a waste of time here.

 

[imagoon]

Not sure if you are serious?

Arcade machines dont exactly use highend hardware. So the cost difference between a thin client and a fixed setup would be minimal at best.

Secondly, all arcade machines got a physical visual design to a game and certain bottons/controls and numbers of players to match.

Thirdly, arcade games almost last forever

Else you could just go to some internet cafe.

I think we reached a point where you _want_ cloud gaming. Try OnLive...oh wait...

I agree here. I would also add that the typical "arcade" is close to dead now also. You really don't need a whole lot more power than an xbox360 for most stand up arcades. I mean look at Golden Tee units, even the brand new ones are basically ~$400 PC's inside. Those things make a mint at the bars also. At that point the thin client would cost more to implement.

 

[imagoon]

www.dictionary.com? Thats a pretty clear sentence I formed. I'm postulating that the way all this is computed could very well be completely different than how we would normally do so on a desktop environment.

Not really, especially in the context of the quote. You confused two people and no one has bother to respond. You could have replied with "I like apples." The sentence would be concise but would not be applicable to the quote in the same way.

 

[kelco]

Not really, especially in the context of the quote. You confused two people and no one has bother to respond. You could have replied with "I like apples." The sentence would be concise but would not be applicable to the quote in the same way.

:whiste: What a constructive addition to the thread. Is this what I have to look forward to in this subforum, because I'll gladly leave if this is what anandtech's reader base is. Jeebus.

 

[Gryz]

I am sorry to say, Kelco, but I agree with the others here.

Stating that "the paradigm of desktop rendering and cloud rendering might not be similar" is too vague for me to answer. The sentence is correct. But isn't that the whole point we are discussing here ? Without stating how you think they might be different, I can not reply. That doesn't mean nobody wants you to have this discussion here. It just means we don't know what to reply to. And we don't agree with things you said.

"I've worked a gameroom, I've been an It administrator and I'm a gamer". But have you been a developer/programmer ? Sometimes I think would help people to get a better understanding of software (and hardware) if they had build something themselves. You start to think like "what is the problem here, what are the different ways to solve the problem".

The problem with programming problems is usually "time versus space".
The issues with networking are usually "bandwidth and delay".

Imho the problem of bandwidth can be solved by throwing money at it. Buy bigger pipes. But the problem of delay can not be solved. Speed of light. I keep repeating.

Have you ever heard gamers complain ? About framerates ? About lag ? About mouse-lag ? They want a smooth experience. No matter what. Not only for FPS games. But also for RPG games. For MMOs. They mess with settings, with V-Sync, with anything they can think of to get a smooth game. With cloud-computing-gaming, you will always introduce extra delays. Delays that will be noticable.

Then you come with the suggestion about usage over a local LAN. That means usage will be much less large-scale. One of the benefits of OnLive is that you can deploy hardware that might be used 12 hours per day, or more. Compared to my gtx680 at home, which is used (for gaming) only a few hours per day at most. When you do cloud-computing on a small scale, you lose the advantage of big numbers and easing out averages. But the complexity stays the same, the management costs are much higher. I don't see the benefit. Besides that, Arcades are dead. And I can't come up with other scenarios where such a scenario makes sense.

Also, cloud-computing is still relatively new. Is it the success that people predicted ? Will it be an even bigger success ? I am not convinced yet. Cloud-computing has advantages, and is here to stay. But I think many companies will realize that their administration and organization is one of their key aspects. And their automation is too important to out-source to cloud-computing. Do you expect banks or stock-brokers to use cloud-computing ? Nope. Too important. I expect many more companies to realize that.

Cloud-computing has its use, but not for everything. And imho, there is no use for cloud-computing in gaming. Demanding games run best on the local machine. For simple games, a browser will do.