Predict: The Next Generation Console Tech

Status
Not open for further replies.
lol, let's just forget all the differences then talk about differences? You're hilarious. There's no point furthering this discussion.

How is it relevant to Nintendo providing overlapping mobile experiences?

What was your point in bringing up differences in mobile/consoles?
 
How is mobile relevant to home consoles? Why do you keep posting off-topic arguments?

It was an example of a games machine which had great success and the following year, introduced the next generation.

It isn't unheard of and wouldn't be the first time.

In fact, since Nintendo shifted to optical media, they have adopted backwards compatibility in the console space as well.

Sony did much the same with ps2/ps3.
 
No, actually it was the DS.

I know. I was playing you. The Atari Flashback also launched in November of that year.

As was already pointed out, you can't compare the mobile and console space directly.

Nintendo has largely went unchallenged in that space so their market dynamic was different. The PSP did well, but I don't think anyone will argue it all affected DS sales.
 
It was an example of a games machine which had great success the year prior and the following year, introduced the next generation.

It isn't unheard of and wouldn't be the first time.
Sure, for mobile space. What does this have to do with the home console space? Quit going off-topic with these tangential examples to support your agenda. It's not going anywhere with your one liner question-arguments.
 
Sony_consoles_shipments.jpg


1998 greatest sales year ever for ps1
1999 ps2 launch

Another example that great existing sales does not preclude a HW producer from launching a nextgen machine.
 
Sony_consoles_shipments.jpg


1998 greatest sales year ever for ps1
1999 ps2 launch

Another example that great existing sales does not preclude a HW producer from launching a nextgen machine.

What does any of this have to do with next gen tech?

Sounds like this tired, and flawed, argument should be in the other thread, not here.
 
What does any of this have to do with next gen tech?

Sounds like this tired, and flawed, argument should be in the other thread, not here.

Predicting nextgen tech is based on timeframe.

When nextgen is expected will dictate what can be fit under the hood in that timeframe.
 
I don't disagree with TheChefO's idea that launching while your current box is strong isn't inherently a bad idea. I do object strongly to the fact that every damn thread he post in turns into "launch in 2012".
Can we please get back to the thread's purpose of exploring the range of hardware options for next-gen? Present suggestions for a 2012 box and a 2013 box and a cheap box and an expensive box and whatever you think will happen, but otherwise everyone stop with the tiresome attempts to predict the business machinations in a hardware tech thread. The in-depth business choices aren't needed to discuss tech options short of picking for a particular design option, and those business topics can be discussed in the general console forum.
 
Don't those business/marketing processes directly influence what type of hardware we'll see? Would you expect a thread on next gen console logistics/marketing to not delve into hardware?
 
Don't those business/marketing processes directly influence what type of hardware we'll see?
Yes, but only which choice is made out of the available options. We don't need to know when MS or Sony is launching to speculate on hardware choices - if we did, this thread wouldn't have started until maybe this year. Instead we can look out what architectures will be out in what timeframes and speculate a 2012 console, or a 2013 console, etc. And most people have been happy talking hardware until recently when the thread was hijacked to talk about the business. A mention of one's theory as to when a box will launch and so one's expectations of hardware is fine, but the hardcore debate of such business options doesn't belong here. Anyone wanting to prove that a console will release in 2012 and so such-and-such hardware will be used needs to take that to another thread.
 
If Chefo's guesstimations (i compromised there :)) turn out to be right. He will be literally fisting humble pie down your throats!

Hahaaa! :p

You'll find me sitting in the corner cringing.
 
Looking at some of these factors and expectations:


  • 230mm^2 = smaller side of last gen GPU footprint which ranged from 230-260mm^2; 10% less area than Barts (6870, 255mm^2)


  • 2.75B transistors = 230mm^2 Barts style GPU with 80% scaling from 40nm to 28nm


  • 700MHz = Less than a Barts 6850 (770MHz, 127W TDP), 6870 (900MHz, 151W), Cayman 6950 (800MHz, 200W), and 6970 (880MHz, 250W). 28nm should bring a solid reduction in power but the increase in transistors is going to scale up power draw. A 6850 is a reasonable 127W considering the 128GB/s of memory bandwidth but a console will need to accommodate the optical drive, HDD, CPU and system memory, etc. With costs (yields/binning) and the RRoD (and YLD) firmly in memory conservative clocks will be likely although the “turbo” features in current GPUs indicates that 700MHz is on the very low end of what should be expected. Comparing a 6790 @ 840MHz 150W max TDP versus a 6850 @ 775MHz 127W max TDP indicates a chip with more functional units and more net performance uses less power than a chip with fewer units at higher frequency.


  • 2100GFLOPs = 80% scaling from 40nm to 28nm minus ~ 10% for space reduction (Barts 255mm^2 to our 230mm^2), ~ 15% for redundancy, and ~ 22% reduction in frequency. GFLOPs may also be hit by the new SIMD+Scalar GCN architecture and DX11.1 overheard as well as additional raster pipelines; this may be higher due to many units not needing to scale and shaders are often easier to pack in. e.g. It is unlikely ROPs will scale from 32 to 64, there may even be a reduction to 24 or 16 ROPs on a console GPU, so this space may be utilized for more Shader units.


  • 76 TMUs, 52.5GT/s = What is a TMU? I am basing this on Barts style TMUs with 56 TMUs at 35% increase of units. For comparison Cayman has 96. A 6870 (56@900MHz) is 50.4GT/s; 6970 (96@880MHz) is 84.5GT/s.


  • 16 ROPs, 11.2GP/s = Or 24. A 6870 is 28.2GP/s (32 ROPs @ 900MHz). When Xenos and RSX shipped they had 8 ROPs when competing PC GPUs had 16. Consoles will have some limiting factors like targeting at most 2GPixel resolutions 1080p (and possibly 2 x 1080 with 3D / 2 player “split” screen) but most games will be 720p 30Hz scaled up to 1080p; there will also be a limiting factor of memory bandwidth. Consoles are about maximizing resources and 32 (or 64) ROPs doesn’t seem like an investment console designers will make when that area could be spent on more shaders.

I think this is on the conservative side. The CPU will be smaller than Xenon IMO (and most certainly CELL) and even this conservative GPU considers the fact more processing will be sent to the GPU

....
Ps- Sony/MS please, one of you, prove me wrong.

I think the above is a reasonable approximation for nextgen 28nm GPU.

Agreed on comments of gpu budget taking some away from cpu budget along with workload.

However, I disagree on the notion that such a GPU would be "weak".

I suppose much of that is due to the intended use by developers. Such a GPU when targeted at 720p and 30fps will provide a sizable bump in visuals. When seeing COD using sub 720p and not suffering in sales, I'm not seeing the devs being convinced to target 1080p 60fps.
 
It is a safe assumption that we will see plenty of 30Hz 720p games next gen. And at that res/framerate even a Barts should produce stunning visuals.

I'm more worried about the amount of RAM. Al seems to think 2GB is likely if I read his posts correctly, and in 2015 this will seem really, really, really low. Like, your cellphone will have way more than that.
 
I'm more worried about the amount of RAM. Al seems to think 2GB is likely if I read his posts correctly, and in 2015 this will seem really, really, really low. Like, your cellphone will have way more than that.

Well, that was assuming GDDR5... Feel free to use 16GB of cellphone memory though. ಠ_ಠ
 
2GB if fast and used efficiently could be plenty.

Personally though, I'd rather they used a larger pool of slower RAM (8gb) and a good chunk (64mb) of EDRAM for a scratch pad for CPU and GPU use to offset the slower lower bandwidth ram.

A larger pool would enable larger persistent worlds, synthesized models/textures/characters/behaviors, and streaming textures could be done more aggressively with such a large pool so there would be less pop-in and preloading areas/levels to counter the relatively slow optical drive.
 
Ok, I want to start a serious line of discussion. Based on the AMD "future GPU" hints let's discuss:

(1) Stacked Memory. When will it be available in volume? What are the specs in regards to Bandwidth, Power, and Cost. What constraints are we looking at in terms of # of modules and bus? What are the providers? What are the pro/con and what are the competing options worth considering?

(2) Silicon Interposer. When will these be available on a mass production scale? What kind of advantages are we going to see from this (wide, low powered busses?) What is the cost of a 500mm^2 "chip" on 90nm and 65nm these days? How much room would one need for a 230mm^2 GPU and 2 dense (stacked?) memory modules? What yield and production issues would AMD be facing? What kind of interconnect would be used to a CPU in such a fabled setup?

And SI addresses a number of bandwidth and PCB trace issues, but are there better options in regards to cost? eDRAM, going stacked memory alone, XDR2 (...), etc.
 
Status
Not open for further replies.
Back
Top