Skip to main content
New Participant
November 13, 2015

Can we establish the real facts about PC hardware for VR?

Hi, VR devs!

As a PC hardware enthusiast, I'm trying to ensure that my PC is optimally powered for VR, and I'm in a position where I give build advice to large numbers of other PC gamers, many of whom want their PCs to be ready for Oculus CV1. Because I'm in this position, I've been getting into lots of debates, mainly related to the Oculus minimum hardware specs and the likelihood of multi-GPU support in the near future. I've spent a lot of time talking to fellow gamers, when really I should be talking to actual developers. In that spirit, I hope that my questions are welcome on this forum. Please let me know if I should post this elsewhere. Trust me when I say that if you can help me settle this debate, you will be helping many other PC gamers as a result.

I'd like to focus this thread on the minimum specs quoted by Oculus: a GTX 970 (or R9 290) and an i5-4590.

Just looking at the numbers, the raw graphical power you need to display game images on a CV1 screen is enormous. Taking into account the 1.4x 'eye buffer' on the 2160x1200 resolution, you get a true rendering resolution of 3024x1680 for a VR headset. (Correct me if I'm wrong about the eye buffer -- I assume Oculus and Vive are the same, and I know Vive has a 1.4x render resolution.)

Trying to target 90 FPS/Hz at 3024x1680 requires roughly 90% as much graphical power as rendering 4K (3840x2160) at 60 FPS/Hz. As Oculus have said themselves, it's also more than 3 times the graphical power required to display a game at 1080p at 60 FPS/Hz.

As everyone probably knows, rendering modern games at 60 FPS with 4K resolution requires a huge amount of graphical power. Looking at a modern PC game like Fallout 4, even a GTX 980 Ti averages only 46 FPS at 4K on Ultra settings. That would translate into about 77 FPS on an Oculus. A GTX 970 averages 33 FPS in Fallout 4 at 4K. That's about 55 FPS on an Oculus, assuming it's directly translatable.

If Fallout 4 were to have launched with VR support, you would need a much more powerful GPU than a GTX 970 in order to play it at 90 FPS on the Oculus. In fact, you'd probably need a multi-GPU system, but from my understanding, multi-GPU support for VR is still under development. (It seems to me that multi-GPU support will absolutely need to work for AAA gaming at 90 FPS to happen.)

My concern is that many gamers are going to be building PCs with GTX 970s on the promise that it will run games at 90 FPS. But based on the math, many new AAA games will need much more power than a GTX 970 to run at high framerates.

When debating this, some people have told me that games designed specifically for Oculus will indeed run at 90 FPS on a GTX 970 because that's the hardware that the developers are targeting. I hope they are right, but I'm still skeptical. You cannot escape the fact that you're trying to pump out 457 million pixels per second (3024x1680x90). Based on my knowledge, a GTX 970 simply cannot perform at that level unless the game is only moderately intensive on graphics. You sacrifice looks, or you sacrifice framerates.

A flagship VR game like EVE Valkyrie looks just as visually impressive as Fallout 4, if not more so. I'll be extremely impressed if gamers are able to achieve 90 FPS with a GTX 970 without significantly dialing down the visual quality of the game. I really, really hope they're able to. But I'm going to remain skeptical until I see it happen.

I imagine two broad classes of games that will be released with VR support:

1. AAA games with really high-end graphics. These will have the resources to optimize for a variety of PC configurations, and will support SLI/CrossFire, because that will be the only way to experience the games at their full potential. These games will require much more graphical power than a GTX 970 can provide.

2. Less expensive games where the focus is on gameplay or story rather than graphics. These games will be less graphically demanding, so it won't matter if they support SLI at all, because even a GTX 970 will be more than enough.

My concern, once again, is that many first-generation adopters are building PCs with a GTX 970 (as suggested) and anticipating that they'll be running the flashiest games at 90 FPS when their Oculus comes out. That's certainly the perception that I see in discussions among the general gaming population.

Can someone shed light on this concern? Am I completely off-base, or will the first generation of games designed specifically for Oculus just have to look very simple graphically compared to today's modern PC games if we want to run them on 970s?

    25 replies

    Up and Coming
    November 28, 2015
    You can see gpu view with the sli vr sample. i run that on a titan x sli. The DK2 refresh at 75Hz, the blue lines are the desktop 60Hz refresh and you can see the associated flip queue.

    An eye take 1.2ms to render, they run in parallel. Then we have an almost 3ms copy, it is definitely higher than expected but match the sample on screen timing report. I saw in the past 10ms copy in 4K, but in AFR, i was supposing that the driver decide to lower the bandwidth priority in favor to other system, as the buffer is only needed 16ms later anyway.

    We also can see a small footer to the frame from the oculus driver that has to run on the same adapter. it is a 0.6ms overhead for my GPU.

    So far from the 11.11ms budget to run at 90fps, nvidia+oculus take 3.3ms, that's quite awful. To note that the next nvidia gpu generation is supposed to have more bandwidth for cross adapter communication, something like 60GB/s because yes, this is the bottle neck right now.

    At the bottom of the screen, you can see the queuing from the app, it is where you will see the SetMaxFrameLatency api from the dxgi device behave, to lower the latency means less queuing and get the risk to have the gpu starve for data at some point, by default, dxgi can let the CPU run ahead up to 3 frames.


    Up and Coming
    November 28, 2015
    I think you are missing one point: In contrast to AFR-SLI not only the GPU-load is reduced but also the CPU-load is pretty much cut in half. Especially on scenes with lots of drawcalls/bindings this can make a huge difference. That is probably why we see the near perfect scaling with VR-SLI (our scenes are usually pretty huge, something around 25Mio visible triangles per frames in several thousand geometries and hundreds of materials).

    Also: I talked about normal stereo rendering for good reason: The sync you do in VR hurts performance a lot. In that case the costs for the copy are relevant, probably the reason why we only see about factor 1.8 performance increase when using the oculus.
    However, if you don´t need to sync (like in normal stereo rendering) the costs are irrelevant. Yes, your latency before you see the final image does increase but since you can do other stuff meanwhile the perceived performance will not be impacted.

    4-way SLI is a completely different story however (won´t even start with 3-way SLI). Guess you would need to come up with something quite different for such a configuration to get a benefit (geometry splitting? depth compositing?). But since you can probably count the number of people with a 4-way SLI system on one hand I would not worry too much about that at the moment.
    Up and Coming
    November 28, 2015
    Oh i missed the not vr stereo info and 1.8 only vr scale, oups, possilble that the driver does something smarter, you know the tool to see by yourself now anyway :)

    The CPU cost is irrelevant since DX12 cut it in crazy proportions, and also allow to implement graphic pipelines that can run far more logic on the gpu side. Yes DX12 does not have the broadcasting, but i do not think it is a problem.

    And 3/4 way sli is a bonus, i work on games that does not target 3D in the first place, the next one will do SLI SFR to lower the latency for sure, each GPU will render a sub rectangle of the view, probably with some overlapping margin in order to do some down sample/blur without a need for copy. If you can do 2, you can do 3 or four :)

    Using this extension will also cut you from tools like renderdoc or vsgd ( maybe nsight will support it one day ), on complex games, losing graphic debug tools is enough of a no go zone.
    Up and Coming
    December 3, 2015
    "kojack" wrote:
    "BalorNG" wrote:
    Otherwise, I think I can expect pretty much 2x scaling with two GPUs due to VR SLI, right?

    VR SLI requires programs to be specifically written for it. No existing rift software would support it. Future stuff might, depending on the engine they use.



    probably requires software Oculus SDK 1.0.0
    facebook.com/AntDX316
    Up and Coming
    December 3, 2015
    VR SLI can be implemented right now by anyone, NVidia has released the SDK for it. It has nothing to do with the Oculus SDK.