PS4 Neo will have polaris too

False. TFlops = Alu/Aggregate Shader speed x 2 ops/clock. ALU speed = # shaders X GPU clock speed. You do not need 56 CUs to get there. 14nm process will allow AMD to increase clocks. 

1. Polaris 10 will have 150W TDP and 2304-2560 shaders clocked at 1266mhz or so. Using the lower end version = 2304 Stream Processors X 1266mhz X 2 ops = 5.8Tflops. Rumored MSRP is $249-299 this summer. 

2. Polaris 10 benchmarks show it's almost as fast as the Fury, while using half the power:

http://videocardz.com/60253/amd-radeon-r9-480-3dmark11-benchmarks

3. By 2017, MS will be able to increase GPU clocks more and add 8GB GDDR5X if they wanted to. Either way, 5-6 TFlops will be achievable with a 120-150W Polaris 10 chips; and these are mid-range 14nm AMD cards for 1st 14nm generation. 

4. Polaris 10's die size is rumored at 232mm2, easily small enough to combine with a Zen CPU. 

A Polaris 10 is not an expensive GPU. It's $300 at retail *this year*, that means manufacturing cost is probably closer to $200, give it another 16 months of price reduction ... and you're probably talking about $150/part price for MS. 

Seems reasonable enough to me for a $399.99 console. GPU isn't going to be the issue here. 

AMD can give MS a Polaris part, the Polaris 10 is 32CUs at 5.5 TFLOP, bump that slightly up to 36 CUs to match the Neo, and you get (tada) just over 6 TFLOPS. That likely is exactly what MS is aiming for. It actually lines up fairly well. 

"Polygon have since shared information that Xbox Scorpio will have approximately 6 Tflops vs. PS4 Neo's 4.2 estimated Tflops"

A 36 CU Polaris 10 would be just over 6.187 TFLOPS. That's probably exactly what MS is using. 

We are all kinda saying the same things. The PS4neo is also rumored to be using polaris too. I just don' t see sony or MS putting in an APU that costs more than whatever they launched with. Especially MS which will mean that again they have they more expensive box. And by that time how far behind will they be? 35M to sonys 70M?

Anyways, we will see i guess. These are all just rumors anyways.

You are quoting retail prices. Add-in board partners and retailers add mark-ups, logistics costs, etc. Buying the chips, GDDR5 directly shouldn't cost more than $80-100 for the small Polaris 10 chip. Besides, AMD will offer a good discount for volumes and to have DX12/GCN Eco-system in place to benefit their PC business via GCN-optimized console ports to PC. Performing well in those titles helps to sell AMD's PC graphics cards. The Division, Need for Speed, Far Cry Primal, Hitman, etc. all run very well on AMD cards because they were optimized for consoles' GCN architectures. For that reason it's in AMD's best interest to get Polaris into both the Neo and XB2. It also costs too much $$$ ($100-120 million) to do a new shrink of existing 28nm parts to 14nm. Might as well use 14nm GPUs already designed for the new node. Since GCN 4.0 (Polaris 10) is backwards compatible with all previous GCN parts, using Polaris 10 ensures both backwards and forward compatibility for XB3/PS5. 

The the biggest issue here is not hardware but software related. Sony already said they will not allow exclusive Neo title as, more or less making the extra hardware only viable to increase frame rates or some graphical effects on what are going to be base PS4 games. Pretty meh concept. For XB2, MS would need to allow developers to make exclusives or give a lot more free room to make much better looking games or otherwise the major gap between PS4/XB1 and XB2, combined with 80-100 million of those "old" consoles userbase will chain developers to optimize for the old, not new consoles. For the strategy to work, MS would need to drop XB1 entirely. Othwrwise, there will be a massive 2-3 year lag between 2017's XB2 and when developers start using that hardware. This is why many are sceptical of the new hardware release strategy. 

Actually.... I'm not quoting retail prices. Yes in previous posts i mentioned retail prices just to out things into perspective. But in the post you wuoyed above, those are BOM prices. 

but lets try somwthimg here.....

I'm saying that at the time of launch, the APU in both the PS4 and XB1 cost under $100. $122 and $132 respectively to be exact (but that's including costs of all the processors that went into the boxes, the APUs alone cost less then $100).

So this is what I will ask..... what kinda APU does anyone here think Sony or MS will get for $120-$200. Using Polaris.

And pls remember, there is a CPU and GPU in an APU. In the case of MS they have to also put in an SDram unless they want to break compatibility with all existing XB1 games. There are other processors inside a console besides the APU. There is memory, cooling, drives and power. And lastly, remember that you can't just up the power of one cponent. everything will go up too. 

Plain no!

1. XB1 GPU is 1.31 TFLOP/s (12 CU's, 853MHz)

2. PS4 GPU is 1.84 TFLOP/s (18 CU's, 800MHz)

3. PS4 Neo is 4.19 TFLOP/s (36 CU's, 911MHz)

So we already see it scales with clock and Compute Units/number of shaders.54% m

Basically with AMD it's shaderunits * 2 * Clock in GHz. And every compute unit has 64 shaders.

So to hit 6GFLOP/s we could go with a little less than 50% more shader units than Neo, a bit less than 50% more clock ora mix of both.

50% more shaders than PS4 would be 48 compute units. That would be more than Hawaii/Grenada has and it doesn't really make sense to cut back Fiji that much. But it's pretty much safe that there will be a smaller Vega version, so we could already have a fit here.

At the same time we do already know that AMD is clocking Polaris much higher than Sony Right now 1266MHz. And we do know that Polaris 10 full build has 40 compute units. Doing the math that would lead us to 54% more GPU power than PS4 Neo being absolutely possible with a complete Polaris 10.

Even a 36 CU Polaris 10 with 1.3 GHz would be enough to reach those 6GFLOP/s. And those clocks don't seem to be an issue with 14/16nm FinFET.

While i actually don't believe we'll see that and Xbox 1.5 would be closer to Neo, both likely coming in 2017, it' not much of an issue. And you really don't need Fiji for that.

I feel like the next Xbox will have to be more powerful because its been shoehorned into a corner and also because if it comes out almost a year after the Neo, it sure as hell can't be weaker.   But neither will be able to do 4k gaming at 60FPS.... without being really expensive and I hope thats not where consoles are going.    In the end, PlayStaion4 will still be on top and that can be avoided,  it has too much of a head start and will probably be cheaper.