Predict: The Next Generation Console Tech

Status
Not open for further replies.
Thanks Prophecy2k and Dr Evil for your answers.

Maybe this time Intel would be willing to offer a better deal, if they feel it's important for them to expand their business into the console realm, pushing Larrabee might make them feel that way.

But how could such a deal look like, considering it is unlikely they will allow the IP to be taken to another foundry? Even if they would allow that it would hardly be economically viable as their CPUs are heavily geared toward their process and intel is basically always one die shrink ahead of the rest.

Is the console market that important to intel that they can make an offer that some of the console manufacturers cannot refuse and take a considerable risk of losing money?
 
I donno how true is this but this site here claims that Sony has choosen PowerVR Series 6 architecture by Imagination Technologies for the next PlayStation console scheduled for 2012 worldwide.This company was the one who made the GPU for DreamCast.

IMGTEC's PowerVR technology uses an advance technique called TBDR which can outperform a competing IMR product from nVidia/ATi by 3-5 fold whilst maintaining equal die size and price point.



source:http://fgnonline.webs.com/
 
Considering that Intel's anti-trust methods involved huge inducements to the likes of Dell etc, they're clearly no strangers to incentivising companies to make sure their chips are looked upon favourably.

Yeah, but Dell could still at any time change to AMD chips, so Dell was not totally dependent on intel. Intel had to keep bribing Dell to keep using their CPUs. Going with Larrabee would not allow that option.
 
@nightshade - nothing more than another flight of fancy from our old friend TEXAN*, I'm afraid. There is no actual basis to that story whatsoever.
 
Is the console market that important to intel that they can make an offer that some of the console manufacturers cannot refuse and take a considerable risk of losing money?

I have no idea, but if their architecture is competitive, then imo opinion it's up to Intel, whether they are in or not, because they can make a lucrative deal with console manufacturers. Even with bad examples in the past, console manufacturers don't necessary have to have the right to outsource the chip manufacturing to other places in order for it to be a good deal.
 
This isn't something either Sony or MS would be designed their next box around though...

Each one would be wanting to get enough bang for buck to 1-up the other on performance by as big a margin as they can muster. Both platform holders will be gunning to be first to market also as that would position them as the primary "next-gen" development platform that the majority of multiplaform developers will lead on.

I'm not sure i understand the relevance of your question?

Given MS are going with ATI and will likely pull out some custom "forward-thinking" 5XXX series-like design, for their GPU and accompany it with a pretty standard off-shelf CPU, what could Sony put out in 2012 that would be able to match or exceed this in terms of performance? - Is that along the line of what you were getting at function?

What I mean is something like this:

Having a more powerful system is in itself irrelevant; what matters is having better versions of popular multiplatform games, or at least not having worse ones. If you have worse games it doesn't matter if it's because your system is technically weaker, or because it's technically t3h @wes0m but in reality games run poorly on it.

There are no(t many) Lazy devs, just real devs that live in the real world. If you build a system that for 95% of developers will produce worse results than its cheaper, older competitors then you've done a terrible job of designed your system. It's your fault, and only your fault.

Having a marginally (or even significantly) more powerful system doesn't matter for your first party exclusive games. The games will be tailored to your hardware, only your hardware and will never be proven to actually run better on the other guys hardware. Uncharted 2 would look about as good and be just as fawned over and be just as safe from potentially looking and running better on another system if the PS3 had Xenon instead of Cell.

If there's going to be a high end battle next gen, and Sony intend to be part of it, what matters most is having better technology for multiplatform games - or at least not worse. Then they can just advertise about having the best system and most people will believe them. Because they are Sony. Probably better to focus on exclusives and cool services that they offer anyway.

MS are in a good position because they can build a system with a powerful but cost effective custom GPU and charge into next gen with software development momentum helping them along. Rage, Cry Engine 3 and Unreal Engine 4 will run well on their next system - they'll make sure of that when they pick their hardware.
 
What I mean is something like this:

Having a more powerful system is in itself irrelevant; what matters is having better versions of popular multiplatform games, or at least not having worse ones. If you have worse games it doesn't matter if it's because your system is technically weaker, or because it's technically t3h @wes0m but in reality games run poorly on it.

There are no(t many) Lazy devs, just real devs that live in the real world. If you build a system that for 95% of developers will produce worse results than its cheaper, older competitors then you've done a terrible job of designed your system. It's your fault, and only your fault.

Having a marginally (or even significantly) more powerful system doesn't matter for your first party exclusive games. The games will be tailored to your hardware, only your hardware and will never be proven to actually run better on the other guys hardware. Uncharted 2 would look about as good and be just as fawned over and be just as safe from potentially looking and running better on another system if the PS3 had Xenon instead of Cell.

If there's going to be a high end battle next gen, and Sony intend to be part of it, what matters most is having better technology for multiplatform games - or at least not worse. Then they can just advertise about having the best system and most people will believe them. Because they are Sony. Probably better to focus on exclusives and cool services that they offer anyway.

MS are in a good position because they can build a system with a powerful but cost effective custom GPU and charge into next gen with software development momentum helping them along. Rage, Cry Engine 3 and Unreal Engine 4 will run well on their next system - they'll make sure of that when they pick their hardware.

Poppycock!!!! I can categorically say with much ocnfidence that U2 would look nowhere near as glorious on an RSX and Xenon!!!

Regardless, your latter point (along with the rather puzzling mindset of many on this forum) is that only MS will be looking to ensure an easily programmable platform for the next generation?!?!?!?

I'm very sure that both Sony, MS and Nintendo (to a lesser extent) will have to maintain this issue as a major point of consideration when designing their next console. Neither Sony nor MS wants to be behind with regards to 3rd party software, and all will be looking to learn from the mistakes of this generation moving forward... including Sony (statements made this year from their big dogs about transitioning from a HW focus to a software focus, help to reassure in this light).

I fully expect the next generation to be completely about the software, with each platform holder investing significantly in software as a means to differentiate their platform. I also expect next-gen's development platforms to be much more mature and developed than they were this gen, at the time the SDK's are released to game developers.

Whether Sony decides to go with a CELL based design, leveraging the experience and tools already developed for the PS3 this time, or not I still expect them to be smarter about their HW design choices for the PS4 in order to ensure their HW doesn't give (lazy...lol) developers the effectual "finger" like the PS3, yet still retains the "long legs" required to maintain an increasing level of visual fidelity in software throughout the next generation.

Sony will almost definately not choose an aged and obsolete GPU this time around, so whether they keep CELL and couple it up with a IMG GPU, ATI GPU, NV GPU or LRB, it will more than likely be intended to maintain it's competitiveness with MS' HW and the majority of DX features at the console's time of release.

Software will be the focus next-gen! Especially with every dev/publisher complaining about increased dev costs spiralling. Both Sony and MS know that by making devs' lives harder and in turn driving up dev times/costs, they'll be putting themselves at a huge huge disadvantage next-gen.

Edit:
I also believe that all parties will be looking to innovate heavily with various motion control/3D/other miscellaneous gimmicks next-gen in order to properly milk the casuals and expand the market like Ninty did this time around. Both software and peripherals are important in this regard, but software again would be the primary means by which to do this effectively... Alas, this is a tech forum ;)
 
Last edited by a moderator:
Poppycock!!!! I can categorically say with much ocnfidence that U2 would look nowhere near as glorious on an RSX and Xenon!!!

It does look great. And possibly even glorious. But I won't fawn over it because it won't run on my PC!

Regardless, your latter point (along with the rather puzzling mindset of many on this forum) is that only MS will be looking to ensure an easily programmable platform for the next generation?!?!?!?

I'm very sure that both Sony, MS and Nintendo (to a lesser extent) will have to maintain this issue as a major point of consideration when designing their next console. Neither Sony nor MS wants to be behind with regards to 3rd party software, and all will be looking to learn from the mistakes of this generation moving forward... including Sony (statements made this year from their big dogs about transitioning from a HW focus to a software focus, help to reassure in this light).

Oh I think Sony will be looking to be as developer and multiplatform-engine friendly as possible this time, and for the kind of reasons you mention.

The reality of making an affordable (to develop and manufacture) system with high performance might make them make compromises though. I wouldn't rule out another appearance from Cell.

Whether Sony decides to go with a CELL based design, leveraging the experience and tools already developed for the PS3 this time, or not I still expect them to be smarter about their HW design choices for the PS4 in order to ensure their HW doesn't give (lazy...lol) developers the effectual "finger" like the PS3, yet still retains the "long legs" required to maintain an increasing level of visual fidelity in software throughout the next generation.

I don't think the "Long legs" thing is anything Sony should worry about. A popular system that captures the core and mass markets will have long legs regardless, and the sheer complexity of writing software and the scope for making better assets means games will continue to improve even if the hardware stays the same.

Sony will almost definately not choose an aged and obsolete GPU this time around, so whether they keep CELL and couple it up with a IMG GPU, ATI GPU, NV GPU or LRB, it will more than likely be intended to maintain it's competitiveness with MS' HW and the majority of DX features at the console's time of release.

I think Sony are still stinging from putting so much into Cell and getting so little out of it, while simultaneously letting themselves get into the situation where they had to use RSX. I don't think they'll let the same thing happen again!
 
So you would like to continue developing with the Cell rather than something the would likely be easier to develop on? Honest question.

I'm much rather develop on a platform where the CPU provides scope to offload heavy lifting onto when your bottlenecked..

PS3's problem w/ respect to the fabled "ease of development" is not the Cell.. It's the RSX...
I say this because it's not hard to program for Cell if you don't need to use the SPUs but with the deficiencies in the RSX it kinda forces you to "need to"...

A PS4 with a better Cell (better PPU(s) + more SPUs) and a much stronger GPU would be an excellent piece of kit that I'm pretty sure most of us could do a heck of alot with from the off, both from leveraging legacy SPU code and investing code into filling up the oodles of GPU processing time you'd have left with most of your (heavy) systems (animation, physics, post processing etc..) already populating the SPUs...
 
@nightshade - nothing more than another flight of fancy from our old friend TEXAN*, I'm afraid. There is no actual basis to that story whatsoever.

Some old Amiga nut like me should do some similar threads about the rebirth of the Hombre chipset.

Sorry, a mod somewhere mentioned the PA-RISC in either this or the end of the Cell thread, so I couldn't resist. :D
 
For a fresh start:

How would a possible 2012 custom Fermi design for the PS4 GPU square up against a mighty little ATI chip around the same timeframe?

Since i know it's not yet 2012 and so no-one could possibly give anywhere near a conclusive answer, but as a point of speculative discussion focussing on the architectural differences between the red & green teams GPU designs going forward, how would one think a comparable chip of 2012 from each company would compare?

Note: I stated a "custom" Fermi design as i understand that Fermi will have alot of supercomputing-relevant DP stuff in there that would be effectively wasted die space in a console GPU. However upon examination of the baseline architectural philosophy behind Fermi, i.e. highly programmable (almost LRB-like) GPU design, I'd like to know if such would be a good fit for a console GPU next gen?
 
For a fresh start:

How would a possible 2012 custom Fermi design for the PS4 GPU square up against a mighty little ATI chip around the same timeframe?

Im sorry to say it but given that Fermi is late and the work on the successor is still ongoing you'd be looking potentially a completely new architecture successor to Fermi and possibly the R1000 series from ATI.
 
After earlier reports about the new GPU for PlayStation 4 Sony is now told that the IBM Power CPU 7 has selected the next generation console the PlayStation 4.

Further reports in English, it is a rumor and not confirmed by Sony.
"We can officially reveal in this world exclusive that SCEI has officially chosen IBM's currently in development POWER7 architecture for it's PlayStation 4 system, currently scheduled for a 2012 worldwide release.
IBM shall POWER7 debut for the server market in the summer of 2010. The PlayStation 4 shall use a cost effective version of the architecture custom designed for Sony's specific needs.
The only information available regarding FGNOnline currently has specifications and performance is that the implementation of the PS4 chip cores shall use 6.8, 24-32MB shared L3 cache, Quad core by threading, and a double precision performance approaching 200GFLOPS.
The CELL chip which powers the PS3 is capable of 15 GFLOPS double precision. This would give the PS4 a performance leap of over 10 fold over it's predecessor.
Interestingly, the primary reason IBM canceled development on the CELL is based PoweXcell 8i because of SCEI's change of stance. The company wanted to move away from the exotic architecture employed for the PS3 to a more traditional architecture favored by third party developers.
As displayed by it's choice of processors, the server based POWER7 and the TBDR based PowerVR 6, SCEI has sights set once again for performance leadership in the next round. "

http://translate.google.com/transla...oor-ibm-power7-cpu-playstation-4/&sl=nl&tl=en




^ I suppose it's plausible. I'd be surprised personally if Sony dumped the Cell architecture..
 
http://translate.google.com/transla...oor-ibm-power7-cpu-playstation-4/&sl=nl&tl=en




^ I suppose it's plausible. I'd be surprised personally if Sony dumped the Cell architecture..

The internet is an amazing breeding ground for compulsive liars.

Please please spend a few seconds doing a reasonableness check before reposting "world exclusive" news from random sites, it would lower the noise level on this board a lot.

Recently it kind of exploded, (thanks to TEXAN) but he is not the only one.
 
Because we all know that game developers cannot live without fast double precision math ;)
Hum... from lately reports it sounds like you're right the truth is that game developers cannot live without energy drink :LOL:

(Cf a eurogamer article about RedLynx studio).
 
If the rumour holds even a shred of weight then they should rip our some of the per core dp float unit silicon and bump up the vector units..

I must admit though..

32 threads, 100GB/s sustained mem bandwidth & 32MB L3 cache (PSP's total sys RAM.. on-die..) does sound rather fetching...
 
Status
Not open for further replies.
Back
Top