Split frame rendering is dubious, it's up to developers to support it. Seeing how many have proper support for decent CFX/SLI, I doubt that'll be meaningful. We'll see.
SFR is important to VR.
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Split frame rendering is dubious, it's up to developers to support it. Seeing how many have proper support for decent CFX/SLI, I doubt that'll be meaningful. We'll see.
As for DX12 performance and asynchronous compute, funny how closer to when it matters nVidia wins in AotS. Let's reserve judgment until release time, shall we? No point speculating on alphas/betas.
We only have two requirements for implementing vendor optimizations: We require that it not be a loss for other hardware implementations, and we require that it doesn't move the engine architecture backward (that is, we are not jeopardizing the future for the present).
So I tried the AMD beta drivers .... and what a surprise .... Crossfire is still not working.
GPU2 usage shows 0% in crossfire
Have you tried the workaround which has been posted several times? The driver doesn't claim to fix CF on its own.
http://www.oxidegames.com/2015/08/16/the-birth-of-a-new-api/
This does not mean that Nvidia has better Async it means that the way Oxide implemented it for Nvidia currently works faster but this has nothing to do with the feature itself which is not the same as it is for AMD at this moment.
and a follow up to thisFrom an API point of view, async compute is a way to provide an implementation with more potential parallelism to exploit. It is pretty analogous to SMT/hyper-threading: the API (multiple threads) are obviously supported on all hardware and depending on the workload and architecture it can increase performance in some cases where the different threads are using different hardware resources. However there is some inherent overhead to multithreading and an architecture that can get high performance with fewer threads (i.e. high IPC) is always preferable from a performance perspective.
When someone says that an architecture does or doesn't support "async compute/shaders" it is already an ambiguous statement (particularly for the latter). All DX12 implementations must support the API (i.e. there is no caps bit for "async compute", because such a thing doesn't really even make sense), although how they implement it under the hood may differ. This is the same as with many other features in the API.
From an architecture point of view, a more well-formed question is "can a given implementation ever be running 3D and compute workloads simultaneously, and at what granularity in hardware?" Gen9 cannot run 3D and compute simultaneously, as we've referenced in our slides. However what that means in practice is entirely workload dependent, and anyone asking the first question should also be asking questions about "how much execution unit idle time is there in workload X/Y/Z", "what is the granularity and overhead of preemption", etc. All of these things - most of all the workload - are relevant when determining how efficiently a given situation maps to a given architecture.
Without that context you're effectively in the space of making claims like "8 cores are always better than 4 cores" (regardless of architecture) because they can run 8 things simultaneously. Hopefully folks on this site understand why that's not particularly useful.
... and if anyone starts talking about numbers of hardware queues and ACEs and whatever else you can pretty safely ignore that as marketing/fanboy nonsense that is just adding more confusion rather than useful information.
Which workaround?
Many have been posted and the results appear to vary per person. Some people use the skyrim profile (this actually just disables xfire), some people report success with the crisis 3 or far cry 4 profile, yet others report it doesn't work at all, or it works somewhat and then frame rates crash.
All the fixes that I have seen posted may increase frame rates however it also creates unreal microstutter effectively making the "fix" useless.
Since consoles are the primary dev target, and they dont have mgpu, it stands to reason that xfire/sli support is an afterthought.
Just they way the world works unfortunately.
Sad, but true. This is pretty much the reason I stick to single GPS systems these days. I'm not holding my breath, but I hope DX12 can go some ways to changing this.
So I tried the AMD beta drivers .... and what a surprise .... Crossfire is still not working.
GPU2 usage shows 0% in crossfire
http://forums.overclockers.co.uk/showthread.php?t=18704111Official Crimson out now!
Fallout 4 Profile is finally here, its absolutely flying on my Crossfire 7970's!!!
Hmm, another single card increase I think guys, pmsl.
Seems to have a profile in the crimson centre but still doesnt want to use my other gpu just checked with afterburner
AMD OWNED
Yes.. There is no crossfire profile. Just tried with default mode and game runs in single card mode.
The OPs second post in that thread reads as follows:
and further down there is this:
The OPs second post in that thread reads as follows:
and further down there is this:
Why are you so mad? Nobody claimed these drivers would fix CF
So .... yeah it did claim to fix CF.Fallout 4 Profile is finally here, its absolutely flying on my Crossfire 7970's!!!
Mad? lol
The post that your posted quote relates stated (in a quote): So .... yeah it did claim to fix CF.
No. Some dude claimed that it did. AMD did not claim it.
I could easily blame the vendors on anything random fools post.
So this will end as another Pieter3dNow Against razor1 thread. always discussing about Async?. so it seems that Async its the only word that Pieter3dNow know to "debate".. man, this is endless I wish I could have the razor1's patience =)
Check carefully if the use of a separate compute command queues really is advantageous
Even for compute tasks that can in theory run in parallel with graphics tasks, the actual scheduling details of the parallel work on the GPU may not generate the results you hope for
Be conscious of which asynchronous compute and graphics workloads can be scheduled together
https://developer.nvidia.com/dx12-dos-and-donts
Not my debate how things work in hardware from Nvidia website ...