Microsoft: Xbone Is 3x As Powerful Than X360 (Phil Spencer Twitter: It Was A Mistake)

[Michael-5]

Pemalite said:

Michael-5 said:

That's why PS3 never really reached it's maximum, it relies heavily on CPU performance, and most people didn't design games which used it. I think now that the PS4 and One mirror PC specs a lot more, people will be able to push the systems better.

I'll look at those links later, gotta go, but even if the GPU takes on CPU tasks, and even if the GPU is the biggest improvement in the PS4 (except RAM) and the most crutial, it's still only one component. If the CPU is sourced from a laptop, and only a few times better then the one in the PS3, a good GPU can't just make that fact dissappear.

Actually, a good GPU can make that difference dissapear, that's the entire point. - Aka. Offloading CPU processing to the GPU, why you keep arguing that point I'll never know. :P

I did manage to dredge up a benchmark comparing PS3-esque GPU to mobile ones:


So from that data, the GPU is is around 3x as fast as the IGP found in Brazos, AMD's slowest IGP currently and still utilising VLIW5.
Thus, proof is in the pudding, the Radeon 7850-level GPU the PS4 uses is vastly superior to the Geforce 7 in the PS3 in every single way, it's no contest, they don't even compete in the same league.
I'm certainly not going to sell my Radeon 7970's and get triple SLI Geforce 7 cards happening as it's a far cry from your stated "Barely any improvement".

Edit: That benchmark is merely rasterization, not compute, that's an entirely different ballgame and would make the Geforce 7 look anemic.

I don't think I said there was "barely any improvement", I said the GFLOPS is 9x as much, but the speed gain isn't as much.

Model	Launch	Codename	Fab (nm)	Transistors (Million)	Die Size (mm2)	Businterface	Memory (MiB)	Clock rate	Config core1	Fillrate	Memory	GFLOPS (Single-precision)	TDP3 (W)	GFLOPS/W (Single-precision)	GFLOPS (Double-precision)	Release Price (USD)
Core (MHz)	Memory (MHz)	Pixel (GP/s)	Texture (GT/s)	Bandwidth (GB/s)	Bus type2	Bus width (bit)	Idle	Max.

Radeon HD 7870 GHz Edition	Mar 19, 2012	Pitcairn XT	2800	212	PCIe 3.0 ×16	2048	1000	1200	1280:80:32	32	80	153.6	GDDR5	256	2560	10	175	14.629	160	$350

Model	Launch	Code name	Fab (nm)	Bus interface	Memory (MiB)	Core clock (MHz)	Memory clock (MHz)	Core config1	Fillrate	Memory
MOperations/s	MPixels/s	MTextels/s	MVertices/s	Bandwidth (GB/s)	Bus type	Bus width (bit)

GeForce 7800 GS	February 2, 2006	G70	110	AGP 8x	256	375	1200	16:8:16:8	6000	3000	6000	562.5	38.4	GDDR3	256

As you can see, the Core clock rate is only a 3x improvement in the PS4. In comparision to the 9x improvement inGFLOPS, a 3x improvment in Clock Speed isn't much. I don;t understand why 2-3 people quoted me and laughed, do I have my facts wrong?

HoloDust said:

Pemalite said:

Michael-5 said:

The new GPU isn't that much faster then the one in PS3. Yes it's more effiecient, but it has to be when the speed barely improves, and the CPU is sourced from a laptop, and something like only 5x as powerful and 1.5x as fast. I know you can't measure brute power by spec alone, but when the majority of components only see 3-5x improvements, and the GPU and RAM alone see greater improvements, it's hard to argue that the PS4 is 9x or more powerful.

I can't beleive I just read that.
A Geforce 7800 class GPU is not that much slower than a Radeon 7850 class GPU?

I know, I was also in "is this guy for real?" state when I've read it.

"Ignorance is bliss" I suppose...

Like I showed above, it's only a 3x improvement. Still big, but not crazy compared to the FLOPS increase of 9x.

I think it's the way I say things, people get offended. PS4 is obviously a pretty powerful system, I'm just saying it's not a huge jump, just big. From PS2 -> PS3 it was like a 25x in performance, now from PS3 ->PS4 it's about 9x, and the Clock Speed is only 3x of a jump. 3x < 9x < 25x, lol.