Can AMD or Intel create a real SOC solution; universal for consoles and PCs?

[a_ump]

OT

SoC's i def feel are possible for the next gen, though i also see it as being possible to get a high end CPU and GPU and later on integrate them into one, more revisions, more releases, more purchases, more revenue. They did it with the 360, the CPU and GPU i believe are in the same package. Look how many revisions/releases it had. Definitely helped them generate more income. I know several people that went out and bought a 360 once it got to its Slim factor, but before they, for some reason, didn't bother.

Makes sense from a marketing POV IMO.

Aren't there already many many threads about tesselation...

 

[crazyeyesreaper]

yes we already mentioned it was off topic about 20 posts ago but i was enjoying the friendly banter

 

[HalfAHertz]

It's interesting that nobody is considering streaming based consoles. Yes, current implementations are lacking mainly due to the limited infrastructure but by the time the next gen consoles are released, speeds and latencies would have improved sufficiently and so would have compression algorithms, hopefully decreasing bandwidth requirement further without the loss of quality. Realistically today you can stream 1080p video with hardware similar to what you can find in most smart-phones(and they're all based on a SoC solution). I'd imagine that 2-3 years from now, you'd be able to do that with a device the size of a match box...Nobody likes loose cables all over the place - wireless is the future

So to summarize: Will next gen consoles implement SoCs? Most certainly yes. Will they be x86 or x64? No. There's no reason for that. In the following years x86-64 will slowly phase out of consumer space entirely and remain only in the server sector.

 

[TheMailMan78]

It's interesting that nobody is considering streaming based consoles. Yes, current implementations are lacking mainly due to the limited infrastructure but by the time the next gen consoles are released, speeds and latencies would have improved sufficiently and so would have compression algorithms, hopefully decreasing bandwidth requirement further without the loss of quality. Realistically today you can stream 1080p video with hardware similar to what you can find in most smart-phones(and they're all based on a SoC solution). I'd imagine that 2-3 years from now, you'd be able to do that with a device the size of a match box...Nobody likes loose cables all over the place - wireless is the future

So to summarize: Will next gen consoles implement SoCs? Most certainly yes. Will they be x86 or x64? No. There's no reason for that. In the following years x86-64 will slowly phase out of consumer space entirely and remain only in the server sector.

Bandwith caps anyone?

 

[lilhasselhoffer]

It's interesting that nobody is considering streaming based consoles. Yes, current implementations are lacking mainly due to the limited infrastructure but by the time the next gen consoles are released, speeds and latencies would have improved sufficiently and so would have compression algorithms, hopefully decreasing bandwidth requirement further without the loss of quality. Realistically today you can stream 1080p video with hardware similar to what you can find in most smart-phones(and they're all based on a SoC solution). I'd imagine that 2-3 years from now, you'd be able to do that with a device the size of a match box...Nobody likes loose cables all over the place - wireless is the future

So to summarize: Will next gen consoles implement SoCs? Most certainly yes. Will they be x86 or x64? No. There's no reason for that. In the following years x86-64 will slowly phase out of consumer space entirely and remain only in the server sector.

1) There are already companies that are trying the streaming game services. It works, assuming you have a good internet connection. Considering the bandwidth capping and service provider tom foolery this is not an option for about 90% of people in the world.

2) SOC is ubiquitous. Crack open your router, modem, or smartphone and you'll very likely find a SOC. The price point for features is great, and the work required to produce the board is minimized. This point is not contestable. What is under contention is whether a SOC can actually perform well enough to cover both graphics, memory, processing, and remain within a budget and thermal envelope. Looking into the near futue, I agree it is very likely if not completely inevitable.

3) Did anyone else find information on the death of x86? Some ananlysists suggest it may die in the next decade because ARM is cheaper, but that's like saying high end restaurants are going to die because fast food has better pricing. It's looking at one aspect, and forming your opinion without considering any other influences. Not to mention that the windows implementation of ARM will not be reverse compatible (according to M$), so the initial investments into the platform will be prohibitively steep. Jumping the gun of the death of x86 is premature, and that's giving it quite a bit of benefit from the doubt.

4) I agree, x86 in gaming consoles is stupid. You have to pay the licensing fee to design for them (Intel), then you have a console that is easier to hack (original xbox becoming cheap servers anyone?), and finally you have no real benefits. It would be great for cross platform development, but it exposes the hardware to too much potential for outside investment at licensing fees. While I can see the value, I recognize that the costs are far too steep right now. Whenever you have something like ARM running both consoles and PCs I see the cross platform environment exploding. Until then, cross platform games are almost soley going to be AAA titles and x86 will only be for PCs.

 

[Benetanegia]

I don't think console makers will do that and in fact I'm sure they will fight as much as they can against that development. Creating a console is not easy, and very few players can enter the market succesfully, making it profitable for those who actually can. With streaming anyone could stream to PC, tablets or smartphones. Said consoles would become just another streaming device, and by redundancy, completely innecessary. Also what would prevent EA, Activision, etc from creating their own service? If their choice is between streaming their games through a service provideed by Sony, M$ and Nintendo (or some others), having to pay a fee or otherwise creating their own service to stream to open platforms like PC or Android devices, I'm sure they would probably create their own.

Also don't current streaming services PAY publishers in order to have the games available rather than what Sony, M$ and Nintendo would want which is to make money from it??

That and bandwidth caps.

 

[Doherz]

Streaming just isn't feasible.
Some markets yes, but globally no.
Just look at the disparity in internet speeds and caps just among the most developed countries? The infrastructure isn't there yet.

In the UK where I live you have two options for speedy internet. Pay for cable, which is vastly overpriced. Or play the postcode lottery for non cable broadband providers. Also if you live in the city of Hull you can't even choose who provides your internet.
Thats just one country and a pretty small one at that, so imagine this on a worldwide scale?

So based on that its could be years if not decades before streaming is feasible for even 25% of the prospective console market, which just isn't profitable.

So the next generation of consoles have to be able to play games locally.

So the question remains when will SOC's be powerful enough for games consoles??

 

[lilhasselhoffer]

So the next generation of consoles have to be able to play games locally.

So the question remains when will SOC's be powerful enough for games consoles??

To the first part, I agree with you fully.

To the second, right now. There are large SOCs that are capable of 1080p gaming right now. For less than the cost of an xbox you can get something like the beagle board. It won't do extremely impressive graphics, but a little bit of work and splicing in components from either Intel or AMD and you could wind up with one heck of a competitor for current gen consoles.
The question is not capability, it's profitibility. At this time the major chip fabs don't have mass production designs for SOCs because, the consumer waters are still being tested. Once one of the giants commits to an SOC (perhaps in the next generation) you'll have a fiscally viable competitor.

 

[Benetanegia]

To the first part, I agree with you fully.

To the second, right now. There are large SOCs that are capable of 1080p gaming right now. For less than the cost of an xbox you can get something like the beagle board. It won't do extremely impressive graphics, but a little bit of work and splicing in components from either Intel or AMD and you could wind up with one heck of a competitor for current gen consoles.

So it's not right now by any means. What we have now in SoCs are able to compete with current gen consoles, but you need much more than that, a whole lot more. XB360 and PS3 were easily 10x times faster than their predecessors, which is what you'd expect 6-8 years later. Upgrading to 2x as powerful systems is what PC gamers do, for what little improvement that increase in performance can offer, console gamers won't move a finger unless the improvement is really noticeable. And same goes for developers. They will never ever create a next-gen tittle unless the platform really makes a difference. A current SoC would only be able to produce the exact same graphics that current consoles do except that instead of at 720p they would do it at 1080p and some AA. Not a single gamer would be willing to pay for a new console for so little in return, especially when they would be playing the exact same games as they would on their XB360/PS3 they already own.

In 2 or 3 years SoCs will be powerful enough to differentiate enough to current consoles, but IMO they will still be far too little and too late. No matter what they say there's more and more people who are discovering the PC as a viable (and superior) gaming platform and in the next years until new consoles arrive, more and more will see the massive difference. Maybe they did not make the jump to PC gaming, nor are willing to do so right now, but everybody is aware of the vast superiority of the platform, and when current consoles launched they were very powerful machines. So if consoles fail to match or surpass the PC at the time of release, very few people will buy them. One of the appealing factors for consoles is that they are better than most PCs when they are announced and first demoed (usually 6-12 months before release), a SoC would miserably fail doing that and people would absolutely loose interest.

 

[Thatguy]

It's interesting that nobody is considering streaming based consoles. Yes, current implementations are lacking mainly due to the limited infrastructure but by the time the next gen consoles are released, speeds and latencies would have improved sufficiently and so would have compression algorithms, hopefully decreasing bandwidth requirement further without the loss of quality. Realistically today you can stream 1080p video with hardware similar to what you can find in most smart-phones(and they're all based on a SoC solution). I'd imagine that 2-3 years from now, you'd be able to do that with a device the size of a match box...Nobody likes loose cables all over the place - wireless is the future

So to summarize: Will next gen consoles implement SoCs? Most certainly yes. Will they be x86 or x64? No. There's no reason for that. In the following years x86-64 will slowly phase out of consumer space entirely and remain only in the server sector.

they are going to go x86 x64 becuase it makes sense to have one set development tools and arm isn't powerful enough for the expectations of the consumer and PPC is pretty much dead with the exception of the limited production in the current xbox, Its a huge software exspense writing code for all those platforms. With AMD and Intel looking to have very powerful SOC's based in x86 and good ease of software tools, it makes ZERO sense to support ppc or arm in a console.

 

[HalfAHertz]

they are going to go x86 x64 becuase it makes sense to have one set development tools and arm isn't powerful enough for the expectations of the consumer and PPC is pretty much dead with the exception of the limited production in the current xbox, Its a huge software exspense writing code for all those platforms. With AMD and Intel looking to have very powerful SOC's based in x86 and good ease of software tools, it makes ZERO sense to support ppc or arm in a console.

Making sense and making the biggest profit are not always one and the same. You write the code only once and then you just compile it for different platforms. Not saying that it's not hard but it's not like you have to reinvent the wheel - if you wrote your code for multi-platform use from the get go, then it's pretty straight forward and there are even automated tools to do that for you.

Saying RiSC is dead is pretty much just as naive as stating that x86 is dead There are a lot more RiSC based devices around you than x86 ones. Phones, cameras, media players, tablets, e-readers and what not. Then there are talks about ARM based netbooks powered by the next gen Tegra platform by Nvidia - a quad core CPU with geforce based graphics being able to compete with atom for less than a fraction of the power. Then you have IBM on the server side, which is a massive behemoth about twice the size of Intel and the majority of their server sales are still RiSC based.

Let's just say that you've had SoCs based on RiSC a lot longer than you've had x86 ones. That's a lot of experience and it's only natural that if the console manufacturers do decide to adopt a SoC solution, it will most probably be based on a RiSC design. What I'm more interested in is if it's going to be a "thin" client or a "fat one" i.e. will it be just a streaming platform or a full-fledged local computing device. Sure there are the benefits of being able to play your games locally but there are also many benefits in having your games streamed - everything is on the server. That means that the hardware can be serviced by professionals (no RRODs ) 24/7; the software will be automated automatically (no annoying update screens) and you'll always be playing the latest patch of the game automatically; cheating possibilities will be decreased further and so on.

Let's face it - companies like Sony and Microsoft and even Nintendo would like nothing more than to have complete control over the platform and to lock it down tighter than Guantanamo bay.

With streaming anyone could stream to PC, tablets or smartphones. Said consoles would become just another streaming device, and by redundancy, completely innecessary. Also what would prevent EA, Activision, etc from creating their own service? If their choice is between streaming their games through a service provideed by Sony, M$ and Nintendo (or some others), having to pay a fee or otherwise creating their own service to stream to open platforms like PC or Android devices, I'm sure they would probably create their own.

Thinking that anyone can stream to the device is a bit naive too. You'd need a tremendous server infrastructure to support the current console populous. That's a pretty hefty investment and I think that only the major console players would be willing to put that kind of dough down on the table. The smaller studios will have to fold or break and once Sony and MS have them under their belt, the bigger players will soon follow to negate the risk of a costly and lengthly market battle.

 

[bostonbuddy]

streaming of games will prob be the gen after next. Maybe streaming of the small arcade games.
Next gen consuls will be the platform to monetize ddl games and media for the masses.
If the next gen consul was either a beefy amd 16core cpu w/ integrated dx11+ graphics + 2 additional beefy amd gpus or an intel 12core cpu w/ integrated dx11+ graphics+ 2 beefy nvidia gpu's, you would have a pretty mean platform that would be able to handle anything you could throw at it for a long time.

 

[Benetanegia]

Thinking that anyone can stream to the device is a bit naive too. You'd need a tremendous server infrastructure to support the current console populous. That's a pretty hefty investment and I think that only the major console players would be willing to put that kind of dough down on the table. The smaller studios will have to fold or break and once Sony and MS have them under their belt, the bigger players will soon follow to negate the risk of a costly and lengthly market battle.

Well I didn't mean anyone as in everyone, but there's a lot of companies which are arguably bigger than Sony, Microsoft and definately Nintendo, at least in the area we are talking about. I'm talking about Google and Facebook for example. Other posibilities:

Intel is not right now prepared (nor is willing to) to enter the console maket as they are simply not what they do, however if providing a gaming sreaming platform service is as easy as creating a lot of servers... hello! Which brings IBM and Samsung to the table too.

We can not rule out Valve. They must have massive ammounts of money when 50% of games sales are digital now and 70% of digital is Steam.

AMD and Nvidia while smaller might want to enter the market too, albeit in a more modest way. Remember you are streaming to common devices, general purpose devices who could connect to any streaming service. Companies offering the streaming service don't necessarily need to offer every single game, partnership or contracts could happen between EA and Nvidia, Activision and AMD or Intel or... you get it. Under such a model even behemoths as Activision and EA could fall down, as developers start doing their thing on their own and talking to streaming services on their own.

And of course you have OnLive and it's competitors who are very modest companies in comparison to the above, but they are there, offering the service already. How easy could be for any of the hardware companies mentioned above to adquire OnLive and create a platform out off nowhere?

The fact is that when it comes to streaming there's plenty of companies which would be much better prepared to fight that battle than Sony or Nintendo, and arguably M$, because the simple fact that they are the hardware suppliers or they already have massive networks in place (MSN means M$ is stronger on this front, but not hardware wise).

 

[Thatguy]

Making sense and making the biggest profit are not always one and the same. You write the code only once and then you just compile it for different platforms. Not saying that it's not hard but it's not like you have to reinvent the wheel - if you wrote your code for multi-platform use from the get go, then it's pretty straight forward and there are even automated tools to do that for you.

.

Its not even remotely that easy to change archtectures. Xbox will likely be x86-64 and it will be equvilent in price to arm and offer more performance and it will be a SOC chip.

 

[EastCoasthandle]

I think with the onset of SoC solutions would also come the option to create a cloud base gaming solution. Would it be easy for MS to buy out either Onlive, Gaikai, Spawn Labs or Otoy? Sure but will they do it? Probably not. But the way I see it, with SoC offering upgrades (we have HDs as upgrades this gen) and the possibility to offer the same games at the same price without having to give you a disc or worry about user bandwidth constraints, to me, sound enticing. Imagine some sort of Xbox consoles, tablets, phones, etc that can directly connected to xbox live service which also provides cloud base service.

 

[Thatguy]

I think with the onset of SoC solutions would also come the option to create a cloud base gaming solution. Would it be easy for MS to buy out either Onlive, Gaikai, Spawn Labs or Otoy? Sure but will they do it? Probably not. But the way I see it, with SoC offering upgrades (we have HDs as upgrades this gen) and the possibility to offer the same games at the same price without having to give you a disc or worry about user bandwidth constraints, to me, sound enticing. Imagine some sort of Xbox consoles, tablets, phones, etc that can directly connected to xbox live service which also provides cloud base service.

Oh god no, the cloud can go suck on lemons. Its my data, I paid for it and I want my right of ownership.

 

[EastCoasthandle]

Oh god no, the cloud can go suck on lemons. Its my data, I paid for it and I want my right of ownership.

I understand that but all you have to do is look at the many posts of people who prefer to download steam games instead of actually owning a hardcopy and you have your answer as to the adaptability to cloud gaming. And who but the publishers/developers would back it when it could put a halt to selling those games and buying used ones. But that's venturing on another topic though.

Anyway, we've seen Rage (a yet to be released title on PC) be touted on handheld devices. How hard would be for them to incorporate it in some sort of cloud base solution and sell it for full price?

 

[Thatguy]

I understand that but all you have to do is look at the many posts of people who prefer to download steam games instead of actually owning a hardcopy and you have your answer as to the adaptability to cloud gaming. And who but the publishers/developers would back it when it could put a halt to selling those games and buying used ones. But that's venturing on another topic though.

Anyway, we've seen Rage (a yet to be released title on PC) be touted on handheld devices. How hard would be for them to incorporate it in some sort of cloud base solution and sell it for full price?

the problem of the cloud is, who owns the data. Last games I bought from steam I made hardopys of, and I don't care about there policy on the matter. Its my game, I purchased it, its mine to keep a hardcopy of.

Just what I want, 3d FPS gaming on a screen small enough to give me eye strain. WTF ? where is the anrgy gamer nerd when ya need him ?

 

[EastCoasthandle]

the problem of the cloud is, who owns the data. Last games I bought from steam I made hardopys of, and I don't care about there policy on the matter. Its my game, I purchased it, its mine to keep a hardcopy of.

Just what I want, 3d FPS gaming on a screen small enough to give me eye strain. WTF ? where is the anrgy gamer nerd when ya need him ?

You get no argument from me about ownership, I understand. But there seems to be so many who just don't care about ownership. All they would have is just the name of the game's title on their 'app' and that's fine enough for them.

 

[EastCoasthandle]

One author thinks that Sony should buy Onlive. Source. But I think the gist of all this is, if I were to guess, is that next gen consoles are going to more then just new hardware. I get the impression there maybe some sort of onlive or steam type of service. But we will see.

 

[TheoneandonlyMrK]

the problem of the cloud is

quite deffinately infrastructure and the gr8 lack of it, their isnt the bandwidth nor the compute power to take on the console gameing comunitys every whim, not this decade maybe 2020- 2030 but not yet

like this thread il have to read thru it tho wow ,some long posts

 

[EastCoasthandle]

I wouldn't think it would be for AAA and other titles but virtual space for indie games. Which could be sold on the:
console
pc
mobile devices
Using HTLM 5 for example. I still think that we may get 1 universal xbox live program that's compatible with PC, mobile and console devices.

 

[TheoneandonlyMrK]

For example, I still think that we may get 1 universal xbox live program that's compatible with PC, mobile and console devices. Using HTLM 5 for example.

fully agree that that will happen just not all cloud based gameing, and deffinately arcade or lower fidelity games as well as high q entertainment functions streamed content etc

 

[FordGT90Concept]

That won't happen until networks are faster and cheaper than computers and I really can't see that ever happening except in corporate environments where the per-computer work load is minimal and bandwidth is high and fast.

 

[seronx]

Supposedly the next PS4 on rumors

Will be using an IBM CPU that is like AMDs Bulldozer
and an AMD GPU

I would say AMD+IBM can and already have made pretty good SoC consoles

IBM and AMD so far based on the same rumors as above are all in the next gen

No intel or Nvidia insight in the next console generation