Post Xbox One Two Scorpio, what should Sony do next? *spawn* (oh, and Nintendo?)

No PS4 Neo at E3, maybe changing the specs?

 

Not remotely possible with a fall 2016 launch date set.

 

Just don't want to cannablise on PS4 and PSVR attention. PSVR is Sony's next big thing. They've invested a crap-load in it and are going to be looking for returns. At the moment it's poised to be the biggest selling complete VR platform and Sony need to carry that through (perhaps to the point where VR crashes and burns in two years' time...).

 

But how likely is it they launch it this fall at this point? I thought we'll get answers this E3 but now we'll have to wait again.

 

This

 

And you win

 

Microsoft1: They are not going to announce the PS4k, maybe we must copy that and talk only about the xbox slim.
Microsoft2: Good idea!
Microsoft3: And you said that this time we should have scheduled the conference after Sony's...

Sony1: They fell for it! Now we can present the new most pouar console evaaaaaa! And cruuuuuush their puny lineup!
Sony2: Hahahahahahahaha!
Sony3: Bwahahahahhaha!
Sony4: Muuahahahahahaha! The world is miiiiiiine!

 

They can't keep secrets like that with all AMD' CEO's winking and orders of "spaghettj al pomodoro with a bit of ps4k reveal this e3 on the side" during busines meetings at restaurants.

 

Come on give AMD some credit. No really, give them some credit. Money is tight.

 

My initial search did not find that specific video, but the discussion I recall was that it had to do with the number of servers contributing to a simulation. I do not remember coming across the part about 20.

The amount of resources that each server brings relative to one console is unclear in my recollection, and I thought this was a multi-player scenario. It would require weighting multiple servers against however many consoles are playing in a simulation.
The challenges in simulation consistency, IO capacity, and persistence in a multiplayer simulation are an area where even if the servers provided less compute power in aggregate, it can still be beneficial.

 

around 9mins onwards. Although most of it is probably worth watching at least once from what I remember.

the bars on the left hand side with the cloud icons are the servers spinning up, may also be in that demonstration where they colour the objects on screen by which server was handing it.

I'm pretty sure that one server was equivalent to 1 xo relative to it doing the physics, not the overall power of the console.

 

Doubtful that they can change the specs at this time, but I'm sure they don't want to announce a 4tflop console right after MS announced a 6tflop console.

Sony will announce NEO at some other event and position it to be an enhanced PS4 with VR capabilities and not a competitor for the more powerful but VR-less Scorpio.

 

most they could do is change the clocks at this stage. Which could give it a nice bump but not tf's worth.

there's been talk of xbox vr games in development. But I expect that they'll just make the rift compatible with the scorpio.
ready made vr unit that's already marketed, well known, right out of the gate

 

Sony upped the ram late in the development, and microsoft increased the clock almost at the last moment

 

neither of them are announcing new consoles at e3. Sony is going to come in very hot if they make 2016.

 

Scorpio is definitely going to support VR, either by supporting headsets generically or by partnering with a specific vendor like Oculus.

 

Oh sure, but my point was that Scorpio isn't being sold solely for the purposes of VR. I'll bet it'll support Oculus, but that will be a completely optional bit. NEO, on the other hand, is going to be sold as the PS4 upgrade for an "optimal" PSVR experience.

 

Oculus in the past has expressed the desire to sell the same headset for many years. Having a rift and a vive I have to say that the rift is almost the perfect vr helm design in terms of egronmics.

For oculus teaming with MS for the xbox would be a god send , even if the deal gave them 0 revenue esp if touch is involved.

1) It allows them to up their production from hundreds of thousands of rifts to millions .

2) It will do the same for touch

3) they will get a ton of support on the pc side for both touch and the headset.


Also even if they want to move to a 2k per eye set up next year or the year after , they can use the same shell and just put in new screens and continue to sell ms the older ones.

 

Funny thing is... this already happened before.

Both were close to being ready to unveil 7th gen, but Hirai surprised everyone with this
http://www.ign.com/articles/2013/01/21/ps4-sony-will-let-microsoft-make-first-move

...and just a fucking week later, this suddenly dropped on official playstation YT channel

Sony could again add clamshell chips, upping the capacity from 8 to 16gigs. On CUH12xx they have moved to eight 1GB gddr5 chips, they have space for another eight.

Or possibly they could add 2-4 new chips if they think they really need them.

 

Why people say fall 2016 for the Neo? I was under the impression it has always been a 2017 product.

At any rate, I think for Sony there is still some room for improvement besides going the obvious overclock route; it only depends how much they're willing to pay AMD...
The often quoted figure of 100M$ for a redesign of the semicustom APU is within reach of Sony and adding more CUs or a wider bus should be quite easy, as the GPU design is inherently modular.

The problem of countering the Scorpio with a simple overclock is that Microsoft could do that too, especially if they are adopting a low clocked 56 CU design as some people speculate. Then, there is also the RAM issue. If Scorpio ships with 12-16GB, multiplatform titles on Sony will always be at a disadvantage as one of the first (and easiest) things to do is lowering the texture quality.