By using this site, you agree to our Privacy Policy and our Terms of Use. Close
Pemalite on 05 August 2017
Edit Thread
fatslob-:O said:
Pemalite said:

I would like for AMD to return to it's small Core strategy that is serving nVidia so well now and what made AMD competitive with the Radeon HD 4000/5000 series.

They did so well during that era.

Even if it means having less hardware features ? 

Who says it will have less hardware features? Smarter architecture design is easily more important than how large and how much you can pack in a chip, it's how nVidia has such a commanding adavntage right now.

fatslob-:O said:

Async Compute is not part of the DX12 specs, it's 'multi-engine' that that is in the DX12 specs but how vendors choose expose it is up to them much like anisotropic filtering.

There is a contradiction here. Haha
Either it is or it isn't.

Even if it's only a part of the "Multi-Engine" that is exposed in Direct X 12. - Then it's still part of the Direct X 12 spec.

fatslob-:O said:

Also DX12 is NOT and open standard, the runtimes, graphics kernel, certification, and the spec is all determined by Microsoft. Much of it goes for Vulkan as well since the spec is determined by the Khronos Group's Architecture Review Board, you can only have an open implementation of Vulkan ...

It is open for everyone to use.

Also, the specs are NOT determined soley by Microsoft.
Like Khronos group, Microsoft seeks input from the entire industry.

fatslob-:O said:


AMD tends to excel at async cause they have a rasterizer bottleneck ... (There's probably very few other reasons for it since AMD highly recommends running a compute shader when doing shadow map rendering which is coincidentally geometry throughput intensive. Nvidia doesn't need async compute all that much cause they have very good triangle throughput.)

Geometry has always been an achilles heel of AMD since the Direct X 11 era bgan.

fatslob-:O said:

AMD can also have their own walled garden such as 'shader intrinsics' and 'rapid packed math' or even 'async compute' so that Nvidia doesn't benefit from these optimizations when these are AMD specific code paths that depend on driver extensions ... (It'd be nice if AMD can get devs to use underestimate conservative rasterization for GPU occlusion culling to gain an optimization advantage since their competitor doesn't offer that hardware feature currently.)

You can't really wall off rapid packed math, asynchronous compute though.

Once nVidia has the hardware that can take advantage of such features, they can place hooks within their drivers to make use of that AMD "path".

fatslob-:O said:

I wonder how many would prefer AMD more than they do now if they just made a bare minimum DX12 videocard but had better performance than their competitor in many current games ? 

I still would opt for AMD over nVidia regardless of performance.
AMD's Eyefinity is superior to nVidia's surround vision, hell Matrox is actually starting to use AMD GPU's for it's workstation GPU's and Multimonitor advantages. (Sad, because Parhelia was a beast back in the day.)

JEMC said:
fatslob-:O said:

Actually quite a few parts of gameworks can run on any DX11 hardware ... 

It's not like Nvidia usually uses their own specific technology in it so it can run on any vendor ... 

And PhysX could run on the CPU, but Nvidia tweaked it so that if you try to do that, the performance would stall. The same goes for their GameWorks things, Nvidia hardware => ok, non-Nvidia hardware => runs very poor.

 

That's why propietary tech is bad for all of us, and we should push for open standards rather than ask for even more propietary software.

Or when you used to be able to pair up an nVidia GPU for PhysX alongside an AMD GPU. But then nVidia disabled that in their drivers, because. Reasons.




www.youtube.com/@Pemalite