Baseless Next Generation Rumors with no Technical Merits [post E3 2019, pre GDC 2020] [XBSX, PS5]

Hidden sauce dreams of nonsense mostly.

 

After XO's 2nd secret GPU believers were sent reeling after Phil Spencer confirmed there was no 2nd secret gpu the believers regained their footing after their messiah MrX stated that it was a clever twist of words on Phil's part and that it really was a secret 3D stacked silicon layer filled with additional shader cores and GPU cache that enormously enhanced the existing GPU.

 

Always funny times prior console releases

 

There's some fudge room on 7nm+, but yes, generally agree. At least it will likely be 7P, which is a 5% boost.

 

Not that it really matters though, unless your grasping for that 0.2TF more or something.

 

Safest bet is both ps5 and scarlett won't have hardware rt (yeah, the 5nm stuff was speculation). I emailed Tom Warren. He won't respond.

I came across this:

Renoir is a complex chip die????

 

The safest bet is that they won’t have hardware RT when both Sony and MS have officially said that they are using hardware accelerated RT?

 

Please stop. There is plenty of official information from Sony, Microsoft and AMD on what to expect from the next-gen systems. This crusade of believing or posting random dribble is pointless, and should be contained in the nonsense thread.

 

An early launch with small quantities of something really extraordinary (even expensive) may really create a crazy hype.....

 

The description of the gpu in project scarlett

According to
https://community.amd.com/community/gaming/blog/2019/06/09/amd-powers-microsoft-project-scarlett

a "Navi" GPU based on next-generation Radeon™ RDNA gaming architecture including hardware-accelerated raytracing

and
http://ir.amd.com/news-releases/new...reports-second-quarter-2019-financial-results
a next generation GPU based on the Radeon RDNA gaming architecture including hardware-accelerated ray tracing

They don't say the same thing. Why is that?

 

And directly from MS's Scarlett video at E3:

Could it be any more clear?

 

So that means rdna1 with RT added, a custom solution then?

 

It seems that the project Scarlett GPU is described both as:

http://ir.amd.com/news-releases/new...reports-second-quarter-2019-financial-results

"a next generation GPU based on the Radeon RDNA gaming architecture including hardware-accelerated ray tracing."

But also as:

https://community.amd.com/community/gaming/blog/2019/06/09/amd-powers-microsoft-project-scarlett

"a 'Navi' GPU based on next-generation Radeon™ RDNA gaming architecture including hardware-accelerated raytracing."

Why is it that they aren't phrased identically?

 

The difference is negligible. Both say it's an RDNA GPU with hardware RT.

 

Exactly.

 

The problem with "hardware-accelerated ray-tracing" is that it could be understood as meaning "gpu-accelerated software ray tracing",

https://blogs.nvidia.com/blog/2019/...ardware-and-software-accelerated-ray-tracing/

 

And huge losses on software.

 

No. Hardware accelerated ray-tracing has hardware dedicated to the task of accelerating ray-tracing beyond what can be achieved without that hardware.

Your link even makes the distinction, calling RT on GPU (compute)...

 

"Real-time ray tracing, however, is possible without dedicated hardware. That’s because — while ray tracing has been around since the 1970s — the real trend is much newer: GPU-accelerated ray tracing with dedicated cores.

The use of GPUs to accelerate ray-tracing algorithms gained fresh momentum last year with the introduction of Microsoft’s DirectX Raytracing (DXR) API. And that’s great news for gamers and creators."

Ultimately DXR is software,

 

What are you talking about? DXR is a software API - it has nothing to do with Hardware RT acceleration. GPUs (APUs/Systems) that don't have RT hardware acceleration will use a software fallback, same as many DX features.

Again, for the third time, 'hardware acceleration' is hardware designs/features incorporated for the purposes of accelerating a task. It's silly to try to imagine otherwise - that'd mean hardware acceleration doesn't exist as a concept. You can do floating-point maths on CPUs without a floating point unit. The FPU was added to accelerate this maths. Clearly a CPU without an FPU isn't accelerating floating-point maths in hardware.