Predict: The Next Generation Console Tech

Status
Not open for further replies.
Is it posible for a console manufacturer who is looking to create a high powered console will use a non standard PSU voltage? Would it be more efficient to use a 24V system to reduce the gauge on the wires supplying power to the system and perhap net a smaller power supply in itself?
 
Is it posible for a console manufacturer who is looking to create a high powered console will use a non standard PSU voltage? Would it be more efficient to use a 24V system to reduce the gauge on the wires supplying power to the system and perhap net a smaller power supply in itself?


The only obstacle would be cost, I would think, considering the higher rated components that would be needed. e.g. capacitor breakdown voltage.
 
The only obstacle would be cost, I would think, considering the higher rated components that would be needed. e.g. capacitor breakdown voltage.

Well, that and the fact that you have to regulate down to 0.9 Volts eventually to power the CPU. You would just be shifting cost away from the power supply wiring and onto the motherboard.
 
In terms of packaging, would it be easier to package the components on the motherboard itself considering the next gen consoles may have a HDD, Optical Drive and PSU built into the chassis?

Im just taking another look at the 24 V thing.

Lastly, how far did talks go with regards to the 24V standard. IIRC there were talks several years ago about changing the standard voltages.
 
Anybody want to try to translate this? An article from Hiroshige Goto

According to a somewhat unreliable source :D, it includes this interesting rumors on PS4:

-Sony surveyed devs about doing a Wii style console and it came back hugely negative. So therefore "Larrabee" high end box was put back as an option, though not a necessity.

-For Larrabee in PS4, conflicting reports, one says it's ruled out, the other that it's still being evaluated.

-If it's a Larrabee system, it will only be Larrabee, no CPU.

-Sony plans to beat Xbox 3 to launch with PS4 regardless -(this seems suspect imo)

-Theres also PSP2 chatter in there and PSP2 is the alleged current gaming priority at Sony. Supposedly similar/small bump in power but with a lot of storage.

Regardless, strange that we've had more chatter about PS4 than Next box..not like the leaky MS..
 
Last edited by a moderator:
-Sony surveyed devs about doing a Wii style console and it came back hugely negative. So therefore "Larrabee" high end box was put back as an option, though not a necessity.

That is quite interesting, specially with many devs reclaiming about the "HD" costs.

Unless the "Wii style console" have a particular meaning, I put this in the strange options.

-
If it's a Larrabee system, it will only be Larrabee, no CPU.

How it is high end with just one processor (supossedely, very hard to work and dev loved cell because of that) if it had more processor why not put a more conventional (intel) CPU too? Dont the pay the licencing by each processor?

-Sony plans to beat Xbox 3 to launch with PS4 regardless -(this seems suspect imo)

That seems quite natural for me.
 
Has anyone been able to confirm the contents of Rangers' link? I would love confirmation on the negative response to the Wii-style console.

That is what worries me most... I will be a sad panda if Microsoft and Sony both take the Wii route...
 
That is quite interesting, specially with many devs reclaiming about the "HD" costs.

Unless the "Wii style console" have a particular meaning, I put this in the strange options.

Well why would next gen go up much from this gen. They already went to HD. Sure with more ram you'd have more textures to produce , but computers have had more ram avalible for that since last gen systems came out anyway.

That seems quite natural for me.

If ms goes 2011 then that means sony has to go before it. 2011 is only a 5 year life for the ps3. To get a meaning full jump on MS (ms luanching in the holiday season) they'd have to launch the first half of the year meaning not even a 5 year life for the ps3. If they go 2010 it be 4 years for the ps3. At that point they would be pulling a sega to me. Heck in 2010 the ps3 might only just be hitting $200.

Now i know that sony will try to keep the ps3 going for as long as possible , but i think it largely rules out the 10 year plan as the system will never reach a mass like the ps2 did before being replaced or even the ps1 unless something huge happens in the next two years.

Ms though launching in 2011 would give them 6 holiday seasons and 5 full years with the 360.

If it's a Larrabee system, it will only be Larrabee, no CPU.

I think then an ibm/ati console form ms would stomp all over it. The single chip would reduce costs and power requirements though.

Regardless, strange that we've had more chatter about PS4 than Next box..not like the leaky MS..

Ms has just gotten better at hiding things. I'm sure its in their best intrest to keep sony guessing as to when the xbox next will come out and i'm sure they are going to be ready to launch with diffrent plans for a 2010-2012 launch. I'm sure once sony announces their plans ms will have an announcement following it.
 
The bit about Larrabee and no CPU is speculation on Goto part. But what he implied is an all Larrabee system. In one old Intel slide, Larrabee can be connected upto 4 way.

The Larrabee core is more potent compare to PowerPC in PS3 and X360. I think it will be bigger also. 16 cores Larrabee might still be 200+mm2 chip on 45 nm. And depending on clockspeed 16 cores won't be enough to run current gen games at 60 fps in 1080p.

Anyway the article said Larrabee is being consider, the rest is pretty much just speculation. Maybe MS and Nintendo is considering Larrabee too for all we know.
 
Anybody want to try to translate this? An article from Hiroshige Goto
-If it's a Larrabee system, it will only be Larrabee, no CPU.

Cell(2?) CPU + Larrabee GPU...hmmm. What would developers say about that combination. Is Larrabee GPU easy to grab after using year and years PC-like ATI/NVidia 3d graphics cards. And backward compatibility would probable be a big issue if anyone still cared it much. I am sure NVIDIA has own interesting offerings available if Sony called them again.
 
this has probably been debated to death, but the Cell 2 SPEs and Larrabee tiles would severely double-team when running all that media centric parallelized code. In essence two chips and architectures would be competing for the same jobs, that would be similar to putting an AMD Fusion and nvidia GPU together in a console.


We can ask ourselves the same question about a Cell 2 with a DX11 GPU off course, but the overlap is not as clear, DX11 GPU won't be based around an extended CPU instruction set as Larrabee is.
Intel blurs the lines. I would actually expect a larrabee tile to be able to run linux in some form while the same can't be said for a cell SPE.
 
Last edited by a moderator:
The bit about Larrabee and no CPU is speculation on Goto part. But what he implied is an all Larrabee system. In one old Intel slide, Larrabee can be connected upto 4 way.

The Larrabee core is more potent compare to PowerPC in PS3 and X360. I think it will be bigger also. 16 cores Larrabee might still be 200+mm2 chip on 45 nm. And depending on clockspeed 16 cores won't be enough to run current gen games at 60 fps in 1080p.

Anyway the article said Larrabee is being consider, the rest is pretty much just speculation. Maybe MS and Nintendo is considering Larrabee too for all we know.
Should be tinier than that or Intel would have cancel the product as as a GPU part (may have still make its way to the gpgpu market if Intel others divisions agree).
Intel hypothesis with larrabee is that they would be able to fit 10 cores within the same die space as a "normal" CPU, it could be likely that they miss their goal but not by that much.
I'm at work right now and I don't have the link to early Intel presentation at hand.
 
I would actually expect a larrabee tile to be able to run linux in some form while the same can't be said for a cell SPE.

There is no reason that an SPE can't run Linux, or any other OS. The seventh, reserved SPE in the PS3 runs the GameOS today. The only requirement to run Linux on an SPE would be a kernel stripped down enough to run in the 256 KiB Local Store, using the 256 MiB of XDR RAM as virtual memory.
 
Should be tinier than that or Intel would have cancel the product as as a GPU part (may have still make its way to the gpgpu market if Intel others divisions agree).
Intel hypothesis with larrabee is that they would be able to fit 10 cores within the same die space as a "normal" CPU, it could be likely that they miss their goal but not by that much.
I'm at work right now and I don't have the link to early Intel presentation at hand.

Yes the paper compare it to Core2Duo. Which is about 107mm2 for 45nm. 16 cores and add the fixed function units and you'll get 200+mm2. That's probably why they'll go straight to 32nm. Larrabee need about 48-64 cores @ 2 GHz to be competitive with AMD and NVDIA future offering. I think if Intel can't secure a console contract, the project will be canned.
 
There is no reason that an SPE can't run Linux, or any other OS. The seventh, reserved SPE in the PS3 runs the GameOS today. The only requirement to run Linux on an SPE would be a kernel stripped down enough to run in the 256 KiB Local Store, using the 256 MiB of XDR RAM as virtual memory.

Ahh.. no. Linux typically requires virtual memory management, memory protection, interrupt handling, more than 256 _K_ilobytes of RAM..

The SPEs are not designed to host operating systems.. they may be adequate for coordinating tasks amongst themselves, but as far as actually driving the whole system, they're not made for it.
 
Ahh.. no. Linux typically requires virtual memory management, memory protection, interrupt handling, more than 256 _K_ilobytes of RAM..

The SPEs are not designed to host operating systems.. they may be adequate for coordinating tasks amongst themselves, but as far as actually driving the whole system, they're not made for it.

Commodore managed to get a real time multi user multi tasking OS Kernel into just 64Kb so it should be pretty easy with 256Kb!
 
Yes the paper compares it to Core2Duo. Which is about 107mm2 for 45nm. 16 cores and add the fixed function units and you'll get 200+mm2. That's probably why they'll go straight to 32nm. Larrabee need about 48-64 cores @ 2 GHz to be competitive with AMD and NVDIA future offering. I think if Intel can't secure a console contract, the project will be canned.
I read the pdf again, you're right Intel hypothesis was that they would be able to fit ten cores within the same die space as a Core 2 duo with 4MB of L2 cache, they also stated that ship would have about the same power consumption (between they don't say per clock which is unlikely).
I found the same references as you here:
http://www.pcgameshardware.com/aid,665558/Intel-Core-i7-Nehalem-CPUs-reviewed/Reviews/
Core i7_45 nm_263 mm²_731 M
Core 2_
45 nm_2x 107 mm²_2x 410 M
Core 2_65 nm_2x 143 mm²_2x 291 M
The aforementioned core2 include more than 4MB of shared L2 cache 6 indeed.
It's not clear if Intel included the fixed function in their calculation but it's not that relevant either the presentation I watched by yesterday night was from the siggraph 2008 and they used the same figures in even earlier presentations. Basically I think that it's a gross approximation and that we should not read too much in it... Anyway I'll do :LOL:

Using Intel gross estimation a 20 cores version (8MB of L2 cache) could take from slightly less than 200mm² to ~270mm².
Best case: 2xCore 2 minus 2MB of cache(should be under 100mm²)
Worst case: nehalem => 4cores+9MB of L2/3 cache+extra logic MC mostly ~270mm²

the 16 cores renditions would land anywhere between 160mm² and 216mm².
But I think that 216mm² would be a failure for Intel or in density or in transistors count per core (~40millions transistors) and the project could be canned.
But I would read too much in Intel waiting for 32nm, Intel has pushed back a lot of projects till 32nm process availability, I think it has more to do with die size and space thus costs.
I makes sense for Intel to avoid pushing a power hungry costly and low yields >300mm² monster.
Intel can wait some more both in the GPU and in the GPGPU market especially as extra time put on the software side on thing will have a huge impact on how the product will be percieved.
I don't think that the project will be canned if Intel doesn't make larrabee the heart of one of the next generation console, it's could still rule the show in the GPGPU realm even if too big/not competitive with Nvidia/ATI on the graphic side. But then it's more an Intel vs Intel decision ie larrabee might not be allow to stealth some market shares from others potentially more profitable Intel divisions.
 
Anybody want to try to translate this?
-Sony surveyed devs about doing a Wii style console and it came back hugely negative. So therefore "Larrabee" high end box was put back as an option, though not a necessity.
Goto writes nothing like that.

He writes SCE visited game developers to ask them about viability of PS4 with Cell 2. In his older article about PS4, his speculation was that SCE actually adopted Cell 2 for PS4, but in this new article he withdrew it. He now speculates SCE just wanted to do preliminary research on game developers' reaction to Cell-powered PS4 before actually designing the console. According to Goto, third-party developers were negative toward PS4 with Cell while first-party developers were positive.

This article is even worse than his older article about PS4, most of it are his speculation pieces as a reaction to the Inquirer Larrabee article without giving a credit to Inquirer. The only interesting part is that PSP2 is just around the corner. Apparently his next article will be 100% speculation on this matter in which he's supposed to tell SCE's fortune, let's hope it's any better than this one.
 
Status
Not open for further replies.
Back
Top