Nextbox: "XBOX Loop"

[Love_In_Rio]

I think is possible for Microsoft to use ARM cores. They are making win 8 for ARM to reach the tablet and smartphone market. If they include this OS core and ARM cores in XBOX you could play your smartphone games in your TV downloading them from a common store, ala Apple TV and Iphones and Ipads.

Besides, we have those rumours from Epic talking about next gen being a bunch of cores. What about project "Ten" being called so for having "Ten" ARM cores?. ( It matches with Kutaragi´s power of two law .

 

[Shifty Geezer]

The Amiga had it right, and did it almost 30 years ago . PCs at the time had no dedicated hardware. By the 90's we saw specialization coming in, with soundblasters and VGA cards.

We also saw PC's super potent CPUs doing all the work of the new 3D games, and Amiga's locked-down hardware couldn't compete, hence it fizzled out.

Nowadays we're so used to these specializations we don't even see them anymore. For instance, you talk about general, but ignore that the GPU is about as specialized and discrete as they come. Every PC nowadays comes with audio hardware (cheapo crap, sure, but it's there and still has SB compatibility deeply buried under it) and the ridiculously specialized GPU. How is that so different from the Copper, Blitter and Paula?

Okay, I'll grant you that, but the talk was of cores for jobs that CPUs are supposed to deal with, not just ancilliary tasks. Workloads have been consolidated where possible. eg. The PPU died very quickly because the same workloads map onto a GPU nicely. If we lose the legacy nomenclature, what we really have now are not Central Processing Unit and Graphics Processing Unit but General Processing Unit and Streaming Maths Processing Unit. Jobs that can be divided between these two resources don't need specialised silicon. Where you still need specific hardware, like a DAC, then you can add a bit of silicon to an existing chip where it naturally fits, like on a GPU.

I suppose in a future console it would be possible for MS to include custom hardware and have it transparently accessible to the devs via DX, so I guess there isn't added awkwardness in trying to write for 5 or 6 different processors, but the whole multiple processor design flies in face of all present developments AFAICS.

 

[liolio]

Any design on the roadmap from ARM itself (A15) wouldn't be competitive. But, there isn't any reason one couldn't build an aggressive, high power processor that implements the ARM instruction set: you would have to implement it yourself rather than licensing the design from ARM. Microsoft has the resources to do this, if they choose. I just have no idea if they will. But, given Microsoft's obvious desire to own all the IP wherever possible, I could see this route being an appealing option from that angle.

Actually if one of the big console market actor were to develop its own high power processor they may as well develop a brand new ISA which doesn't bear the burden of compatibility. Especially Microsoft, they have few legacy code on ARM (as they threw most of it with winphone 7). They may as well develop their own low processors based on this ISA. They won't because it's not their business and I'm not sure they want to invest in the proper human resources to do so. For high performance ARM cores, you make it sound like developing such cores would be a piece of cake, most likely it's not the only companies that have that kind of experience are Intel, AMD and IBM, it's their core business (or was at some point) and they went through a trial/error process doing so with sucky implementation from time to time. So assuming they do their best it's the proof that it's not trivial at all.

The claims I've seen indicates that nvidia intends to take this route with Project Denver: it is claimed to be much more of a PC-level part (in both power consumption and performance) despite being an ARM device.

Well Nvidia is claiming PC level of performances right now with their tegra3 which is imho a joke.
For Denver I would be pretty conservative, any decent GPU will push the power budget into PC-level.

 

[liolio]

Are there examples of a similar cost single-die CPU/GPU combo besting a discreet CPU and discreet GPU of similar costs (power; also remember that a single die chip won't have the same footprint due to defects, power/cooling, etc)? While significant bandwidth is a good thing, I am not sure a single die solution due to communication has really shown to be a big win over having a significantly larger chip (e.g. 50% larger GPU).

I guess llano is a good example as I'm not sure it's possible to put together a discrete CPU a discrete GPU that would beat it significantly. I'm not sure the comparison is fair either, on to discrete parts you pay the various margin (along the chain from the foundry to the seller) twice, but you still need a beefier motherboard whereas (assuming we don't take "upgradability" in account the APU doesn't need one Pci express x8 or x16. The yield will be lower for the APU but taking in account all the other factors (margins assembly costs, etc. paid twice) I guess it depends on how big is the chip.

We can't really measure the improvement that comes with fast and low latency communication between the CPU and the GPU because the only products that offer this are late Intel CPU. Point is we've read comment here and there from a lot of developers that want this low latency communication to make better use of the GPU (Carmack if memory serves right among others even on this board (as early as 2006 "will cpu move to OoO" or T. Sweeney and A. Richard in the interview they gave to semiaccurate).

 

[glw]

Actually if one of the big console market actor were to develop its own high power processor they may as well develop a brand new ISA which doesn't bear the burden of compatibility.

I've seen a few hints that Microsoft was involved in the definition of ARMv8, so they might consider ARMv8 to be their "brand new ISA".

 

[AlNom]

We can't really measure the improvement that comes with fast and low latency communication between the CPU and the GPU because the only products that offer this are late Intel CPU. Point is we've read comment here and there from a lot of developers that want this low latency communication to make better use of the GPU (Carmack if memory serves right among others even on this board (as early as 2006 "will cpu move to OoO" or T. Sweeney and A. Richard in the interview they gave to semiaccurate).

Well, fwiw, it's one reason why the intel GPU has any decent speed at all (access to cache). The problem will be seeing how this scales up to a beefier core because you'll want a lot more cache.

 

[kagemaru]

After the massive success of X360's ATI GPU vs PS3's Nvidia GPU why would MS switch to Nvidia?
Just because Nvidia can make ARM cores?
Please, anyone can make ARM cores. That's the whole idea.
If Apple can build a custom SOC and slap a ARM+PowerVR, I see nothing difficult about slapping N ARM cores + ATI/AMD GPU.
And then if you want to fit the rumors call the Compute Shaders specialized Physics, Sound, AI "cores".

As someone who'll be working on these next systems, what do you think about MS (or Sony) switching to ARM? Would you prefer they stick with the Power architecture? Apologies if you really can't answer this question, I'm just trying to get a sense to how an ARM equipped console would perform next gen.

 

[Arwin]

I would be inclined to look at chips in terms of power to performance ratios. In that respect, ARM proves itself in the mobile space, and if the budget is simply how many cores can I fit into a 200W budget, then ARM should be interesting. Versus a similar good performer such as the Cell's SPEs, I think ARM has the additional advantage in that with the success of smartphones, development and design with ARM architectures are comparatively well known and can give you a huge head-start in development. With ARM now also being a target Windows platform, there's even more advantages in shared R&D for Microsoft right there.

But even then it will be just a small part of the next-gen puzzle, one-third of the ingredients at the utmost I would guess .

 

[AzBat]

I think some may be overlooking one possibility: the device may be a mobile device & not the same o hardcore home gaming device. It might just be something like Shifty has been dreaming about: a tablet.

Tommy McClain

 

[AlNom]

I think some may be overlooking one possibility: the device may be a mobile device & not the same o hardcore home gaming device. It might just be something like Shifty has been dreaming about: a tablet.

Tommy McClain

I hope they market it better than Zune and, you know, release it in more than one country.

 

[fehu]

It might just be something like Shifty has been dreaming about: a tablet.

VADE RETRO!

 

[AzBat]

I hope they market it better than Zune and, you know, release it in more than one country.

I'm sure if it's a flagship Xbox device you'll see it in the same places the 360 is available. Unless of course they decide Japan is not worth it. LOL

VADE RETRO!

That's just being closed-minded. I think Microsoft can improve on the Wii U design & still be a superior device. Execution maybe another matter though.

Tommy McClain

 

[bgassassin]

I think Loop is a dedicated set-top box and Ten is the real successor.

 

[AzBat]

I think Loop is a dedicated set-top box and Ten is the real successor.

If Paul Thurrott is right & Project Ten is coming out next year, maybe it's the other way around? Or maybe somebody got their project/code-names mixed up?

Tommy McClain

 

[Shifty Geezer]

I think some may be overlooking one possibility: the device may be a mobile device & not the same o hardcore home gaming device. It might just be something like Shifty has been dreaming about: a tablet.

Wouldn't you think it strange if someone didn't release a tabler-console? It strikes me as the most obvious product for someone wanting to compete in a less direct fashion than yet-another-box-under-the-TV. If we don't see a tablet console in two years, I'll lose all faith in the world's product designers!

 

[Proelite]

Heterogeneous architecture with different CPUs for different tasks (AI, physics, sound etc.) is laughable.

More likely it will be a powerful SoC with 4GB of DDR3 and maybe some EDRAM or whatever. Perhaps the EDRAM with be a separate die as it is on the 360.

My suggestion (which is worthless) would be a tiny but very capable dual or tri core CPU and a big, flexible DX11+ GPU (think GTX460) on one die (or at least the same package). This would enable some very interesting possibilities and be perfectly feasible even on 40nm. Would maybe draw around 175-200W.

When I see games like Uncharted 3 or GT5 running on PS3, I have to think such a console would be capable of absolutely mind-blowing visuals.

Tell me more about about a 28nm 175-200W next gen console.

 

[fehu]

That's just being closed-minded. I think Microsoft can improve on the Wii U design & still be a superior device. Execution maybe another matter though.

Tommy McClain

then they can sell it as a tea tray an release a real controller for real games

 

[Sonic]

Wouldn't you think it strange if someone didn't release a tabler-console? It strikes me as the most obvious product for someone wanting to compete in a less direct fashion than yet-another-box-under-the-TV. If we don't see a tablet console in two years, I'll lose all faith in the world's product designers!

I wouldn't think it strange at all. Now, if you just mean a portable gaming system that is essentially a tablet with extra controls then yea I can see that happening. If you're talking about the tablet replacing the actual box that sits in the living room then nope, don't see it happening. And the reason I say this is that the tablets need to be complimentary products to the home consoles. You can pack in more punch and power in a $300 box than you can a $300 tablet. I would deem it foolish for any of the console manufacturers to abandon the box that sits in the living room concept as that part of the market is really growing now. We can have on the go, and in the living room.

It would be pretty bad ass of Microsoft to release a tablet gaming system, and the next gen Xbox at the same time. Give the name Xbox brand to both and I'm sure it would do well. It would be cool for cross platform titles, well cross MS platform, to purchase one and have it for both. The possibilities of this are quite intriguing. I would like to see next gen launch of tablet gaming PC and home system at the same time. Might cost upwards of $5 - 600 to get both, but people would end up buying it, especially if Live came standard the Tablet.

 

[homerdog]

The ignorance about AI/physics aside, why is a heterogeneous architecture so laughable?

The ignorance about dedicated AI/physics processors is precisely what I find laughable.

 

[homerdog]

Tell me more about about a 28nm 175-200W next gen console.

Isn't that in line with the original Xbox 360? Seems like a reasonable target, and I'm sure they learned their lesson about inadequate cooling.

Maybe they will go the Wii route but I think that's very unlikely.

Also, the hypothetical console I described was on 40nm. On 28nm it would draw a good bit less.