Maximum PC: Futuremark today announced that the Android version of 3DMark is now available to download, giving Android device owners another benchmark at their disposal. Several prominent technology firms provided input into the benchmark's design, including Imagination Technologies, Intel, Broadcom, Nvidia, Qualcomm, and others. One of the goals was to make it so that scores could be compared across platforms, and apparently Futuremark delivered.
Put your PC through some rigorous testing to see if it's up to the demands of VR.
Asynchronous Compute is a feature that has been in the spotlight for a while. As we’ve seen, more and more developers are taking advantage of it via their DX12 implementations in their game engines. And from the looks of it, this feature will stay relevant for a very long time.
Its a waste of time for people to use Futuremark when it comes to GCN being as it won't take advantage of what's being offered.
@Herbalistic1d ago (Edited 1d ago )
[Its a waste of time for people to use Futuremark when it comes to GCN being as it won't take advantage of what's being offered.]
Agreed, this is where your benchmark can suffer in certain situations of how to show a performance gain or loss in a certain compute dynamic added or subtracted or not even used inside your benchmark.
And since GCN was designed on a Hardware level to use such functionality as Asynchronous Compute in situations where a game or other application use such functionality. By not using such included into your Benchmark programs
You may not be able to accurately show a result if such was used.
Now not all benchmarks are going to be able to benchmark every compute situation ..LMAO..that's why there is more focused benchmark tools.
But atleast there is an opening for benchmarking tools to give you an idea using multiple benchmark tools to gather that compute data based on different Compute run tasks..to give you a result.
AMD definitely has the leg up with Async. We still haven't seen that mysterious Async driver for AotS Nvidia promised when they were caught red handed. It just what it is.
Nvidia cards are built for replacing in 2 years max.
AMD cards shows long legs and are future proof. they might last you much more than 2 years easily.
The word "futureproof" in regards to PC hardware is just silly. I guess it goes back to what tou want to do. I could buy a $400 laptop for my mom that will do everything she wants, email, word processor, facebook, etc. You could say that laptop would be futureproof for her, no?
I had an R9 295x2 but I scraped it for 2 980tis, sole reason, i could hook the 980tis up to a 4k tv, but not the r9 295x2. AMD finally addressed that, but the lack of HDMI 2.0 in the 200 series was some solid bs.
I haven't looked at benchmarks in a while, but I recall my 980tis crushed the r9 fury x in some games, and vice versa in others. All depends on the engine, game, and dev patnership with amd/nvidia.
As far as futureproofing goes, we know there is no such thing. Not if you want the best visuals available when these new games launch. 4k is so pretty that i am willing to upgrade frequently to keep up.
Guess you dont know Nvidia Pascal Gpus support Async too!Amd is actually making new strides but dont expect Nvidia to stand still,actually even without proper Async at the moment 90% of the game benchs made with Async on Nvidia and Amd puts Nvidia gpus having better framerates than Amd.The gains are bigger on Amd Gpus with Async but Nvidia Gpus hardware still pulls ahead of Amd even with lesser gains from Async.In the future we will see but has i said Nvidia is not going to stay still regarding better optimization of there own Gpus to work better with Async.
Revisiting The Term "Asynchronous Compute"
Now Nvidia has been claiming that they support Async Compute since Maxwell. Several developers have spoken out and told everyone countless times that Nvidia GPUs cannot support Async Compute, but that isn’t stopping Nvidia from having presentations on their “Enhanced Async Compute” with Pascal. So basically Nvidia created and actually showed “Enhanced Async Compute” while using DX11. As hilarious as that may sound, Nvidia is serious. They still have GPUs to sell and they must stay competitive, even if they have a large share of the market already.
Nvidia “Enhanced Async Compute” is basically a newer and better form of preemption. The issue with preemption is that it is not asynchronous so therefore it cannot be async compute. Preemption is not parallel at all and can cause overhead issues. Preemption basically causes data to stop for more important or higher priority data that must make it to the GPU immediately. This still causes delays and stops the flow of data to the GPU. So you will end up with the GPU waiting for important data and waiting again for normal data. Context switches must occur and this only cause more overhead issues, this is what DX12 and Vulkan is attempting to remove.
http://www.overclock-and-ga...
Not really, it's still task switching..very fast task switching none the Less it's not the same as AMD's..since they are not the same type of AS...
Your results are going to be off
To put it another way, Nvidia is basically emulating Async through software. Async compute is a hardware solution handled by the Aces in an AMD GPU, something Nvidia GPUs currently lack.
Only thing future proof on a PC is a hefty power supply.
This is good. We need AMD to stay relevant and get some sales.
Futuremark recently released their Time Spy DX12 demo and benchmark. People have been waiting for this benchmark to release for some time now. There has been some controversy surrounding this benchmark. This article takes a look at what going on behind the scenes. Apparently AMD Radeon GPUs aren’t being fully utilized.
