Its that extra generosity of Microsoft and not to even toot their own horn that will swing this thing back around just like the BC.

Not necessarily.  The Xbox One has a stronger CPU, the PS4 a stronger GPU.  This addresses the GPU side of the Xbox One, so it's addressing a significant bottleneck.  The Eurogamer test suggested that Hitman on the Xbox One S slightly outperformed the PS4 version, and they described Hitman as an ideal test case.  *shrug*  It's only one data point, so it remains to be seen.  The real proof will come from games like Gears 4 that are being designed with the Xbox One S in mind.  They've already confirmed the framerate is higher on the Xbox One S, seemingly even with HDR turned on (if I understood the interview correctly).

Yes... ideal to them...

When they and even MS say the differences is neglible and that they where looking at the best case scenarios for improvements you know how much this was shoehorned.

1 or 2 games out of 100's of others that dont. Also Hitman doesnt "always" run better on xbox one, just in 1 small section of the game.

Usually XB1 has less FPS or lower image quality / resolution than the PS4.

A 7% increased clock rate of the gpu, on the Xb1 slim, isnt going to put it ahead of the PS4, or even come close in like 99% of the games out there.

And yet they used a SSD, meh. Just compare them as standard.

Not surprising, it's a mor optimized and stable console.
Excellent news for the games that struggled to maintain a locked 30

Hitman is an 'ideal case test' because of its fairly unique set-up (separate sections clearly push the GPU and CPU, optional frame-rate cap, and no graphical variations), but it's not representative of the average 8th gen game. If developers treated the CPU and GPU equally we would see manifestations of the X1's CPU advantage more often, but given the PS4 and X1 themselves are GPU focused, it's somewhat rare (almost all 8th gen titles are GPU focused, in line with the hardware they're built for). The X1's CPU advantage could be as large as the PS4's GPU advantage, and the latter would still be more significant.

Anyway, while it's nice MS are making the effort (and developers will certainly appreciate it), as DF mention, it'll likely be fairly inconsequential in most instances. Many people don't even consider the standard PS4's 40%~ GPU advantage significant.

I am curious...... so all it took to support 4K mdeia playback was a 7% boost in GPU, a bump in EDram bandwidth and probably the inclusion of HDMI 2.0?

And all inclusions still putting it all round below the PS4 in performance. So does this mean that the only thing preventing the PS4 from doing HDR and 4k media playback in HDMI 2.0?

And another thing, there has been a lot of talk about the existence of the Neo or even the scorpio being due to an inability to shrink the APUs found in the PS4/XB1 down from 28nm to 14/16nm.

But here we see a 16nm APU in the XB1s? Interesting........

The 7% boost or higher clock speed and memory bandwidth was needed to absorb the extra cost of rendering to a HDR output buffer, 10 bit instead of 8 bit.

HMDI 1.4 also supports 10 bit (and 12 bit color) however not HDR 10, the new format used by 4K tvs for HDR. The ps4 and original XBox One could already output games in 10 bit (and 4K 30fps), yet without the extra brightness range that HDR allows. HDMI 2.0 is needed for HDR. (And HDCP 2.0 DRM for video content)

The 20nm planar die shrink failed. XBox One S is now using 16nm finfet which keeps the cost up for now. The release of the 500GB model is likely delayed due to very low to no profit margins, get the hardcore to buy the profitable 2GB model first. Perhaps the slight increase in clock speed is also to compensate for any low level differences between planar and finfet architecture, but that's pure speculation on my part.

Anyway it's good news for the price of the Neo. Without the esram its gpu shouldn't be much more expensive than the 16nm redesign in the XBox One S. Maybe even cheaper as it's a standard component. Neo is partly due to the die shrink problems, we should have cheaper slims already and the relatively higher cost of the delayed slims worked in favor of a mid gen refresh.

Yes... a very good article indeed... we deny 40-50% GPU difference makes any real impact, but a 7% increase in the GPU makes all the difference in the world... and cherry picking exemples (and being clear that even then the gains were small) is a clear show that this guy was trying too hard.